Data analysis and relationship modeling in R. R-analysis, or the acceptability of criterion-based approaches Data analysis in the r language assignments

Random Forest is one of my favorite data mining algorithms. Firstly, it is incredibly versatile; it can be used to solve both regression and classification problems. Search for anomalies and select predictors. Secondly, this is an algorithm that is really difficult to apply incorrectly. Simply because, unlike other algorithms, it has few customizable parameters. And it is also surprisingly simple in nature. And at the same time, it is amazingly accurate.

What is the idea behind such a wonderful algorithm? The idea is simple: let's say we have some very weak algorithm, say . If we do a lot different models using this weak algorithm and averaging the result of their predictions, the final result will be significantly better. This is what is called ensemble learning in action. The Random Forest algorithm is therefore called “Random Forest”; for the received data, it creates many decision trees and then averages the result of their predictions. An important point There is an element of chance in the creation of each tree. After all, it is clear that if we create many identical trees, then the result of their averaging will have the accuracy of one tree.

How does it work? Let's assume we have some input data. Each column corresponds to some parameter, each row corresponds to some data element.

We can choose randomly, from the entire data set a certain number of columns and rows and build a decision tree based on them.

Thursday, May 10, 2012

Thursday, January 12, 2012

This all started back in April 2011 when I had a phone interview with Zynga. Then it all seemed like some kind of game unrelated to reality and I could not even imagine what it would lead to. In June 2011, Zynga came to Moscow and conducted a series of interviews, about 60 candidates who passed a telephone interview were considered and about 15 people were selected from them (I don’t know the exact number, some later changed their minds, others immediately refused). The interview turned out to be surprisingly simple. No programming problems, no tricky questions about the shape of hatches, mostly testing your ability to chat. And knowledge, in my opinion, was assessed only superficially.

And then the rigmarole began. First we waited for the results, then the offer, then the approval of the LCA, then the approval of the visa petition, then documents from the USA, then the queue at the embassy, ​​then additional check, then a visa. At times it seemed to me that I was ready to give up everything and score. At times I doubted whether we needed this America, after all, Russia is not bad either. The whole process took about six months, in the end, in mid-December we received visas and began to prepare for departure.

Monday was my first working day in a new place. The office has all the conditions for not only working, but also living. Breakfasts, lunches and dinners from our own chefs, a lot of varied food stuffed in every corner, a gym, massage and even a hairdresser. All this is completely free for employees. Many people commute to work by bicycle, and several rooms are equipped for storing vehicles. In general, I have never seen anything like this in Russia. However, everything has its price; we were immediately warned that we would have to work a lot. What “a lot” is, by their standards, is not very clear to me.

I hope, however, that despite the amount of work, in the foreseeable future I will be able to resume blogging and, perhaps, tell something about American life and working as a programmer in America. Wait and see. In the meantime, I wish everyone a Happy New Year and Christmas and see you again!

Result<- NULL for(i in (1:length(divs[,1]))){ d <- divs if (d$Divs>0)( try(( quotes<- getSymbols(d$Symbol, src="Finam", from="2010-01-01", auto.assign=FALSE) if (!is.nan(quotes)){ price <- Cl(quotes) if (length(price)>0)(dd<- d$Divs result <- rbind(result, data.frame(d$Symbol, d$Name, d$RegistryDate, as.numeric(dd)/as.numeric(price), stringsAsFactors=FALSE)) } } }, silent=TRUE) } } colnames(result) <- c("Symbol", "Name", "RegistryDate", "Divs") result

Introduction

First of all, let's discuss terminology. We are talking about an area that in Western literature is called Data Mining, and is often translated into Russian as “data analysis.” The term is not entirely appropriate, since the word “analysis” in mathematics is quite familiar, has an established meaning and is included in the names of many classical sections: mathematical analysis, functional analysis, convex analysis, non-standard analysis, multidimensional complex analysis, discrete analysis, stochastic analysis, quantum analysis etc. In all of these areas of science, a mathematical apparatus is studied, which is based on some fundamental results and allows one to solve problems in these areas. In data analysis the situation is much more complicated. This is, first of all, an applied science in which there is no mathematical apparatus, in the sense that there is no finite set of basic facts from which it follows how to solve problems. Many problems are “individual”, and now more and more new classes of problems are appearing, for which it is necessary to develop a mathematical apparatus. An even greater role here is played by the fact that data analysis is a relatively new direction in science.

Next, we need to explain what “data analysis” is. I called it an “area,” but an area of ​​what? This is where things get interesting because this is not just a field of science. A true analyst solves, first of all, applied problems and is focused on practice. In addition, data must be analyzed in economics, biology, sociology, psychology, etc. Solution

new tasks, as I already said, require the invention of new techniques (these are not always theories, but also techniques, methods, etc.), so some say that data analysis is also an art and craft.

IN In applied areas, the most important thing is practice! It is impossible to imagine a surgeon who has not performed a single operation. Actually, this is not a surgeon at all. Also, a data analyst cannot do without solving real applied problems. The more such problems you solve on your own, the more qualified specialists you will become.

Firstly, data analysis is practice, practice and more practice. We need to solve real problems, many, from different areas. Because, for example, the classification of signals and texts are two completely different areas. Specialists who can easily build an engine diagnostic algorithm based on sensor signals may not be able to create a simple spam filter for emails. But it is very desirable to acquire basic skills when working with different objects: signals, texts, images, graphs, feature descriptions, etc. In addition, this will allow you to choose tasks to your liking.

Secondly, it is important to choose your training courses and mentors wisely.

IN In principle, you can learn everything yourself. After all, we are not dealing with an area where there is some secrets passed from mouth to mouth. On the contrary, there are many competent training courses, source codes for programs and data. In addition, it is very useful when several people solve one problem in parallel. The fact is that when solving such problems you have to deal with very specific programming. Let's say your algorithm

gave 89% correct answers. Question: is it a lot or a little? If it’s not enough, then what’s the matter: did you program the algorithm incorrectly, chose the wrong algorithm parameters, or the algorithm itself is bad and not suitable for solving this problem? If the work is duplicated, then errors in the program and incorrect parameters can be quickly found. And if it is duplicated by a specialist, then issues of assessing the result and acceptability of the model are also resolved quickly.

Third, it is useful to remember that data analysis takes a long time to complete.

Statistics

Data Analysis in R

1. Variables

IN R, like all other programming languages, has variables. What is a variable? Essentially, this is the address with which we can find some data that we store in memory.

Variables consist of a left and a right side, separated by an assignment operator. In R, the assignment operator is the construction “<-”, если название переменной находится слева, а значение, которое сохраняется в памяти - справа, и она аналогична “=” в других языках программирования. В отличии от других языков программирования, хранимое значение может находиться слева от оператора присваивания, а имя переменной - справа. В таком случае, как можно догадаться, оператор присваивания примет конструкцию следующего вида: “->”.

IN Depending on the stored data, variables can be of various types: integer, real, string. For example:

my.var1<- 42 my.var2 <- 35.25

In this case, the variable my.var1 will be of integer type, and my.var2 will be of real type.

Just like in other programming languages, you can perform various arithmetic operations on variables.

my.var1 + my.var2 - 12

my.var3<- my.var1^2 + my.var2^2

In addition to arithmetic operations, you can perform logical operations, that is, comparison operations.

my.var3 > 200 my.var3 > 3009 my.var1 == my.var2 my.var1 != my.var2 my.var3 >= 200 my.var3<= 200

The result of a logical operation will be a true (TRUE) or false (FALSE) statement. You can also perform logical operations not only between a variable with a certain value, but also with another variable.

my.new.var<- my.var1 == my.var2

Zazimko Valentina Lentevna Ph.D., Art. Lecturer at the Department of Economic Analysis of the Federal State Budgetary Educational Institution of Higher Professional Education "Kuban State Agrarian University"

The traditional approach to analyzing the financial situation is based on the general concept of “equilibrium of systems”, borrowed from countries with market economies (Figure 1).

Figure 1 — Methodology for analyzing financial condition, corresponding to the Western concept of “equilibrium” of the system

Meanwhile, such problems as: 1) the inconsistency of some methodological approaches with the conditions of the Russian specifics of doing business; 2) failure to take into account the social nature of the agricultural sector of the economy in Russia (when differentiating approaches to assessment depending on the sectoral affiliation of organizations); 3) analysis of the main factors influencing business performance using statistical analysis; 4) structuring the methodology for analyzing financial condition (at least in order to restore linguistic justice); 5) the correspondence of financial analysis to the modern needs of economic entities and the ambiguous interpretation of borrowed economic categories have been studied with insufficient completeness.

The main direction for improving the methodology for analyzing the financial condition of an organization should be to take into account:

The existing political climate and government approach to assessing economic phenomena, processes, and business results;

Features of the legislative regulation of the preparation of financial statements (this particularly concerns the revision of approaches to assessing the solvency of an organization);

The sectoral structure of the property of an economic entity;

Modern parameters for assessing business efficiency.

The purpose of analyzing the financial condition of an organization is an objective assessment of the financial situation and prospects for its development, taking into account the current situation in the industry in a specific time interval corresponding to the general political and economic strategy in relation to the object of study.

The agrarian transformations of the modern era in the history of Russia are deep and significant: since the second half of 2005, the Government of the Russian Federation has significantly intensified its interest in agriculture, initiating, among others, the national project “Development of the Agro-Industrial Complex”; At the end of 2006, the Federal Law “On the Development of Agriculture” was adopted. The state policy of supporting agriculture provides for incentives to attract loans on the terms of subsidizing interest payments. The weakening of the financial independence of the Companies, as a consequence of the measures taken, according to generally accepted approaches to determining the financial condition, is assessed negatively. According to the estimates of domestic economists, who recognize the shortcomings of the existing methodology for calculating indicators of the financial condition of agricultural producers, used, including by arbitration courts (Table 1), there would not be so many bankrupt farms in the country.

Table 1. Fragment of calculation of coefficients for classifying agricultural producers to the groups of financial stability of the debtor

The study of the financial condition of the organization must comply with the concept of consistency. The methodology for analyzing the financial condition of the organization, at the same time, appears in the form of an agreed sequence, which allows us to state the fact of restoring the linguistic validity of the term “methodology”. It consists of six main stages, the general block diagram is shown in Figure 2.

Figure 2 — Flowchart for the implementation of stages of analysis of the financial condition of agricultural organizations

Gathering information consists of compiling a list of questions and obtaining relevant data from the organization being studied and from other sources. The study of the operating conditions of systems should become a preliminary stage of analysis, which is due to the task of an indispensable synthesis of internal and external factors, which arises taking into account the peculiarities of the evolution of economic analysis in Russia, described above. Thus, for agricultural organizations, it is specific to study the geographical, weather and climatic conditions of the business of the analyzed subject. Structuring the initial information should involve compiling data slices that should be included in the information base for analyzing the financial condition of the organization with its main characteristics: industry, business scale, and others.

At the next stage, in the generated array of information, it is necessary to highlight the indicators that are the most important criteria for performance. Many academic analysts, both foreign and Russian, place profitability indicators above other indicators. Thus, E. Altman, in his well-known five-factor “Z-model” for determining the probability of potential bankruptcy, presented two out of five factors as profitability indicators. The importance of profitability indicators is also reflected in the “Golden Rule of Economics”, which states that the growth rate of balance sheet profit must exceed the growth rate of revenue from product sales, and the growth rate of sales must exceed the growth rate of assets.

The criterion for identifying phases in the traditional life cycle schedule is also the profitability indicator (y-axis in Figure 3).

Figure 3 - Organizational life cycle

In combination with absolute financial performance indicators, the key indicators of the activity of an agricultural organization are: gross output at current selling prices, revenue and profit (loss) from the sale of products (works, services), profit (loss) of the reporting year, net profit (loss) , operating capital turnover ratio, return on equity, return on operating capital.

The system of indicators proposed for the purpose of analyzing the financial condition of business entities in the agricultural sector of the economy was tested using the example of actual data from JSC Agrofirm Kavkaz in the Krasnodar Territory. The organization occupies far from the last place in the ranking of the three hundred largest and most efficient agricultural companies based on the results of 2003-2007, included in the Agro-300 club.

Figure 4 - Dynamics of financial performance indicators of CJSC Agrofirm Kavkaz

Analysis of absolute financial performance indicators indicates the development and growth of the company (Figure 4). Thus, steady dynamics in the indicated direction is typical for indicators of gross output (+ 39%), revenue from sales of products (+ 43.9%), as well as the final financial result of activities (+ 16.8%). Among the factors that positively influenced the dynamics of indicators, one can name an increase in the volume of produced and marketable crop products - primarily grain (by 3.4%), sugar beets (13.9%), sunflower (47.9%) and milk (9 ,9 %). The return on operating capital for the reporting period increased compared to the base period, which proves the high efficiency of the joint-stock Company.

In order to identify significant factors influencing the level of business efficiency, a correlation and regression analysis of the business efficiency of 46 agricultural organizations in the central zone of the Krasnodar Territory was carried out. The level of return on equity (in percent) is taken as the effective indicator (y), calculated as the ratio of the net profit (loss) of the reporting year and the average annual balance of equity capital. The choice of this particular indicator is explained by its excessive demand by external users of financial statements as an indicator characterizing not only the efficiency of a business, but also its riskiness, strategic prospects for solvency and the quality of business management. Key indicators-factors that potentially influence the degree of return on equity were selected for analysis; the search and calculation of these factors can be carried out on the basis of public financial statements. These are: x 1 - share of equity in the balance sheet currency, %; x 2 - ratio between debt and equity capital (financial leverage ratio); x 3 - share of liquid funds in assets, %; x 4 – asset turnover ratio (resource productivity).

Analysis of paired correlation coefficients showed that there is a direct and fairly close connection between return on equity and the ratio of debt and equity capital, according to the Chaddock scale, which confirms the statement that the search for a rational ratio between debt and equity sources of financing is a clear path to increasing effectiveness of the latter. The inverse average relationship between the performance indicator and the share of equity in the balance sheet currency (Tables 2 and 3) indicates that the return on equity in modern conditions increases if its share in total capital decreases. At the same time, there is a direct average connection between return on equity and the share of liquid funds in assets, and a direct weak connection between it (profitability) and return on assets.

Table 2. Matrix of paired correlation coefficients of the four-factor multiple regression equation

Analysis of β-coefficients indicates that the weakest influence on the change in return on equity is exerted by the share of equity in the balance sheet currency, and the strongest is the ratio between debt and equity capital. Moreover, precisely according to the second characteristic, the studied population of agricultural organizations is extremely heterogeneous. In addition, this set is heterogeneous in terms of return on equity, the share of equity in the balance sheet currency and the share of liquid funds in assets, which indicates a different level of organization of production and financial activities and its efficiency in farms.

Table 3. General characteristics of return on equity and selected factors, 2006

The multiple regression equation obtained as a result of the solution has the form:

y = -12.454-0.164x 1 +0.688x 2 +0.905x 3 +39.335x 4. (1)

The positive value of the coefficient at x 2 is evidence that with rational farming methods and a normal ratio of return on assets and interest on interest paid on borrowed sources of financing, the profitability of own resources should increase.

Table 4. General results of the four-factor regression model assessment

The relationship between return on equity and all factors included in the model is close (multiple correlation coefficient R = 0.901) and statistically significant (Table 4). Moreover, the linear equation explains 81.2% of the variation in return on equity. The rest is due to random unaccounted factors.

In practice, to calculate the level of business efficiency of agricultural producers and ways to improve it, the main factors and the degree of their influence on the performance indicator are identified. It has been determined that the return on equity of the studied population of agricultural organizations: decreases with an increase in the share of equity in the structure of sources of financing (return on equity increases only up to a certain level of equity and begins to decline with a further increase in its share in the structure of the balance sheet); increases with an increase in the financial leverage ratio, which reflects the ratio of debt and equity capital and characterizes the dependence of profit on the structure of sources of financing, which is possible with a preferential tax burden and support for farms from the Government of the Russian Federation; has growing dynamics with an increase in the share of liquid assets in the structure of the organization’s property, which is logical in the light of the implementation of settlement and payment discipline, and is a consequence of the growth of the organization’s business activity, manifested in an increase in income (revenue) from the sale of agricultural products and other types of activities (priority of the marketing strategy activities of the organization); increases with the level of use of the organization's own assets (a priority task of the organization's financial management).

From here, it becomes possible to form the right vector for increasing the business efficiency of agricultural organizations through the use of clear mechanisms that contribute to its growth. In the most general form, such mechanisms are: 1) a reasonable determination of sources of financing the organization’s activities; 2) increasing the efficiency of using the organization’s resources based on the stabilization of mutual settlements and the system of settlement and payment discipline; 3) improvement of the production management system.

A study of the dynamics of the return on equity capital of agricultural organizations depending on the actual level of the share of equity capital in the structure of sources of financing showed that the highest value of the efficiency indicator for the use of equity capital was recorded at the level of equity capital in the range from 44 to 58%. With further growth of equity capital in the structure of sources, a decrease in profitability is observed (Figure 5).

Figure 5 — Dynamics of return on equity depending on the share of equity in the capital structure

Studying the impact of an organization's financial strategy regarding the use of borrowed funds continues the described sequence.

The developed methodology for determining the rational ratio of borrowed and equity funds in connection with the return on equity capital and preferential lending to agricultural organizations takes on an acceptable place here.

From the entire set of relative indicators of financial stability, we propose to calculate the coefficient of financial independence (Equity to Total Assets), which characterizes the ongoing policy in the field of financing and reflects the share of equity capital in the structure of sources of property, and the ratio of debt and equity capital (financial leverage ratio, or “leverage of financial leverage”), characterizing the degree of risk of the organization.

Capital structure ratios characterize the degree of protection of creditors and investors from possible non-payment of debts and provide practically no information about the economic potential of the organization. The described problem is “solved” by an indicator characterizing the dependence of profit on expenses associated with the structure of sources of financing the organization’s activities - the “financial leverage effect”.

EGF = (1-Neskh) (CRa -PK) x (ZK/SK), (2)

where EFR is the effect of financial leverage, which consists in an increase in the return on equity ratio, %; Neskh - the rate of the unified agricultural tax, expressed as a decimal fraction; CR - gross return on assets ratio, %; PC - the average amount of interest on a loan paid by an organization for the use of borrowed capital,%; ZK - the average amount of borrowed capital used by the organization; SK is the average amount of the organization's equity capital.

Formula (2) was obtained by taking into account the peculiarities of the formation of data in the financial statements of Russian organizations, as well as the taxation of agricultural producers: 1) instead of the entire amount of capital used, in our opinion, the amount of the organization's accounts payable should be subtracted from its value; 2) “the amount of gross profit excluding the cost of paying interest on a loan” was replaced by the indicator “profit from the sale of products (works, services)”; 3) income tax, the payment of which is carried out under the general taxation regime, is not considered by the author as a factor influencing the magnitude of the effect: in accordance with current legislation, agricultural producers pay a single agricultural tax, which was introduced into the formula.

Table 5. Dynamics of financial stability indicators of CJSC Agrofirm Kavkaz

So, the share of borrowed capital in relation to equity in CJSC Agrofirma Kavkaz at the end of 2006, according to Table 5, amounted to 52.8%, which is 42.1 percentage points. greater than the base year level. An increase in the share of borrowed capital in the liability structure of the balance sheet indicates a transition from conservative to moderate financial policy; and although this is associated with a weakening of the autonomy of the business entity, under certain conditions, this can lead to an increase in return on equity. It should be noted that the degree of business activity of agricultural producers is not so high for the implementation of such a financing policy in the future, which means that the consequences of the changes should be carefully studied and a rational decision should be made.

The results of calculations to determine the effect of financial leverage for CJSC Agrofirma Kavkaz (Table 6) indicate its positive dynamics: the value in 2006 was 2.5%, which is 3.3 percentage points. greater than the base year level. Consequently, CJSC Agrofirma Kavkaz, having formed its assets by 65% ​​from its own funds and 35% from borrowed capital, increased its return on equity by 2.5%, all other things being equal, due to the fact that credit resources it pays taking into account the policy of preferential lending to agricultural producers pursued by the Government of the Russian Federation, and the return on total capital is 16.2%. Factor analysis of the model of the effect of financial leverage showed that in the current conditions it is profitable to use borrowed funds in the organization's turnover, since the consequence of this is an increase in the efficiency of using equity capital. This means that by attracting borrowed resources, the analyzed organization can increase its own capital, provided that the return on invested capital exceeds the price of attracted resources.

Table 6. Mechanism of formation of the effect of financial leverage

To determine the limits of growth of financial leverage, one should use the model developed by the French scientists J. Conan and M. Golder. The explanation for this is the composition of the criteria, which is most adapted to the requirements of constructing domestic financial statements. The lower the value of the estimated indicator, the lower the likelihood of delays in payments by the company. The actual values ​​of the criteria, calculated according to the data of CJSC Agrofirm Kavkaz, are presented in Table 7.

Table 7. Assessment of the probability of payment delays of Agrofirm Kavkaz CJSC

Calculations show that the probability of a company delaying payments is very small, however, the dynamics of the integral indicator tends to zero, which means that the level of solvency in the future is under threat. This wave is justified against the backdrop of an increase in the amount of borrowed funds and debt servicing costs. In order to prevent possible difficulties, operational monitoring of settlement and payment discipline is necessary.

In order to synchronize positive and negative cash flows, operational solvency management is necessary. The authors of the study are categorically against the use of liquidity ratios as indicators of solvency due to their contradiction with the accounting requirement of going concern. The degree of solvency, in our opinion, depends on the filling of financial performance indicators with real money. The use of offset transactions in settlements and the replacement of cash with receivables creates a threat to the organization’s ability to meet its current obligations.

Currently, not enough attention is paid to cash flow analysis. Meanwhile, this is the most non-contradictory method that allows us to monitor the degree of sufficiency of funds to cover short-term obligations. Endovitsky D.A. suggests comparing net cash flow from current activities with profit from sales. A negative net cash flow, while there is a profit from sales, will indicate that the formation of working capital requires large financial investments. This situation may lead to insolvency. Reasons: low profitability of sales, high costs for the formation of working capital.

Table 8. The ratio of net cash flow and profit from sales, thousand rubles.

The net cash flow from current activities at CJSC Agrofirma Kavkaz is positive, however, in more detail the adequacy of cash flows to finance working capital will be demonstrated by factor analysis (formula 3):

, (3)

where Dptd is cash inflow from current activities, thousand rubles, OK is working capital, thousand rubles; Dotd - outflow of funds from current activities, thousand rubles. Performance indicator ( Kdost1) in a given relationship characterizes the organization’s ability to finance working capital, shows the sufficiency of cash inflows to cover the costs associated with financing working capital. The recommended value of the indicator should be at least 1.

1. Impact of changes in the net cash inflow ratio for current activities: . (4)

2. The impact of changes in the outflow of funds per one ruble of working capital: . (5)

Table 9. Data for factor analysis of the coefficient of adequacy of cash receipts for financing working capital, thousand rubles.

Thus, the positive change in the cash flow adequacy ratio for the analyzed period (+0.148) is due to an increase in the outflow of funds from current activities to cover working capital. The ratio was negatively affected by the faster growth rate of cash outflows than the growth rate of cash inflows.

According to CJSC Agrofirma Kavkaz, the ratio of cash inflow and outflow for current activities in the reporting period was 1.018, while the dynamics of the coefficient was negative - a decrease of 0.076. However, this does not mean a lack of funds to cover short-term obligations. The cash flow adequacy ratio to cover short-term liabilities is very acceptable both in previous and in the reporting periods (0.966, 4.216 and 2.780, respectively).

Regular monitoring of the current state of funds

Figure 6 — Stages of analyzing the solvency of an agricultural organization

The next step is to evaluate the quality of profit (formula 4):

, (4)

Where NPV- net cash flow for all types of activities, thousand rubles, PE - net profit, thousand rubles.

If, based on the results of its activities, an organization has a persistent negative net cash flow, this may lead to financial insolvency caused by an actual decrease in resources and a decrease in the economic potential of the organization. In the analyzed situation, as can be seen from Table 9, the organization received a net profit, while for each ruble of profit there are 3 rubles of the balanced result of comparing the inflow and outflow of funds. The study of the possibilities of assessing the solvency of an agricultural organization made it possible to formulate an analysis plan presented in Figure 7.

The results of the study are fully based on the realities of the work of agricultural organizations. This resolves the problem of the lack of industry specificity in existing financial analysis methods. The practical significance of the study is that, based on the developed methodology for agricultural organizations, the basis for the formation of a rational financial policy in the transforming economic situation of the rural industry is proposed. Using the recommended methodology will allow you to more accurately measure the level of financial risk and develop a more effective mechanism for managing it in order to improve the performance of business activities.

R-analysis, or the acceptability of criteria-based approaches in assessing the financial condition of agricultural organizations

In the current economic conditions, the main emphasis in the activities of financial services of commercial enterprises is focused on the operational monitoring of indicators of the financial condition of the organization. In this case, priority is given to relative indicators that characterize the relationship between reporting data that carries this or that information. In terminological terms, the method of analyzing a company's activities based on the described approach is called R-analysis, or analysis of financial ratios.

The set of coefficients within an individual business entity depends on the strategy and goals that it wants to achieve. In this case, the coefficients that should be calculated are identified and their standard values ​​are established. This work is usually carried out as part of a management accounting, budgeting or balanced scorecard project. “If a set of indicators is taken from a textbook on finance,” note practicing analysts, “such financial analysis will not bring any benefit to the enterprise” /10/.

Meanwhile, certain indicators relating to aspects of an organization’s financing of its activities have developed traditionally and are included in all methodological algorithms, including those regulated by law.

We are talking about the following indicators:

I. Liquidity Ratios

Liquidity indicators characterize the company's ability to satisfy the claims of holders of short-term debt obligations.

1. Absolute liquidity ratio

Shows what share of short-term debt obligations can be covered by cash and cash equivalents in the form of marketable securities and deposits, that is, almost completely liquid assets.

2. Quick ratio (Acid test ratio, Quick ratio)

The ratio of the most liquid part of current assets (cash, accounts receivable, short-term financial investments) to short-term liabilities. It is usually recommended that the value of this indicator be greater than 1. However, real values ​​for Russian enterprises are rarely more than 0.7 - 0.8, which is considered acceptable.

3. Current ratio (Current Ratio)

It is calculated as the quotient of current assets divided by short-term liabilities and shows whether the enterprise has enough funds that can be used to pay off short-term liabilities. According to international practice, liquidity ratio values ​​should range from one to two (sometimes up to three). The lower limit is due to the fact that working capital must be at least sufficient to pay off short-term obligations, otherwise the company will be at risk of bankruptcy. An excess of current assets over short-term liabilities by more than three times is also undesirable, since it may indicate an irrational asset structure.

Calculated using the formula:

II. Gearing ratios - Capital structure indicators (financial stability ratios)

Capital structure indicators reflect the ratio of equity and borrowed funds in the company's sources of financing, that is, they characterize the degree of financial independence of the company from creditors. This is an important characteristic of enterprise sustainability. To assess the capital structure, the coefficient of financial independence (Equity to Total Assets), which characterizes the firm’s dependence on external loans, is most often used. The lower the ratio, the more loans the company has, the higher the risk of insolvency. A low value of the ratio also reflects the potential danger of a cash shortage for the enterprise. The interpretation of this indicator depends on many factors: the average level of this ratio in other industries, the company’s access to additional debt sources of financing, and the characteristics of current production activities.

Calculated using the formula:

Other indicators, such as: Profitability ratios - Profitability ratios, Activity ratios - Business activity ratios, Investment ratios - Investment criteria, will not be given within the framework of this article for reasons of disclosing the issue raised while condensing the material.

The main thing when conducting financial analysis is not the calculation of indicators, but the ability to interpret the results obtained. The conclusions, however, do not differ in their breadth of scope: the main conceptual approach is based on a comparison of the data obtained with the standards established within the framework of the traditional approach. The traditional approach is understood as a set of methods, tools and technologies used to collect, process and interpret (interpret) data on the company’s economic activities.

Although the main contribution to the theory and practice of financial analysis was made by economists from countries with developed market economies, it is necessary to recall the works of the Soviet economist of the 20s N. Blatov, which outlined advanced concepts and methods of analysis for their time: comparative analytical balance, distribution coefficients, coordination coefficients, etc.

An interesting point is the borrowing and, to a certain extent, interpretation of the “extreme values” of analytical coefficients characterizing solvency and financial stability, with their comprehensive distribution.

Thus, in one of the sections of Y.V. Sokolov’s work, written jointly with V.V. Kovalev, we find a description of the interpretation of Western accounting and analytical practice to Russian specifics. At the same time, information is provided on the financial condition of ten large joint-stock companies in Russia based on the results of work in 1907 and 1908:

“JSC “Caucasus and Mercury” (shipping company), Bogorodsko-Glukhovskaya manufactory, Firm “Provodnik” (rubber and telegraph production), Partnership M.S. Kuznetsova (production of porcelain products), Russian Electrical Society "Westinghouse", JSC Russian Electrical Plants "Siemens and Gallskoye", Singer Company, JSC Maltsov Plants, Bryansk Rail-Rolling, Ironworks and Mechanical Plants (JSC), Society of Putilov Plants "/2 , With. 280/.

A limited list of coefficients is calculated (their list is given above). The average values ​​of the coefficients calculated based on the given sample (the criterion for grouping enterprises is not specified) are compared with “world” standards. When their proximity was detected, a conclusion was made that these values ​​are acceptable in relation to the current situation in the country in the structure of assets and the sources of their coverage /11/.

To this day, there are a number of contradictions, to ignore which, in our opinion, means to remain silent about the main thing.

Let us turn to the instructions (recommendations) of ministries and other federal executive authorities on the aspect of methodological approaches to analyzing financial condition in the context of the coefficients given in them. Among these, the most significant are the methods presented in the documents below:

1. Methodological provisions for assessing the financial condition of enterprises and establishing an unsatisfactory balance sheet structure, approved by order of the Federal Administration for Insolvency (Bankruptcy) of Enterprises under the State Property of Russia dated August 12, 1994 No. 31-r.

3. The reporting procedure for heads of federal state unitary enterprises and representatives of the Russian Federation in the management bodies of open joint-stock companies, approved by Decree of the Government of the Russian Federation of October 4, 1999 No. 1116.

4. Guidelines for conducting an analysis of the financial condition of organizations, approved by order of the Federal Service of Russia for Financial Recovery and Bankruptcy (hereinafter referred to as the FSFR) dated January 23, 2001 No. 16.

5. Rules for conducting financial analysis by the arbitration manager. Approved by Decree of the Government of the Russian Federation dated June 25, 2003 No. 367. These rules, in accordance with Federal Law dated October 26, 2002 No. 127 FZ “On Insolvency (Bankruptcy)”, define the principles and conditions for the arbitration manager to conduct financial analysis, as well as the composition of information , used in this case.

6. Instructions on the procedure for drawing up and presenting financial statements, approved by Order of the Ministry of Finance of Russia dated July 22, 2003 No. 67n.

7. Decree of the Government of the Russian Federation of January 30, 2003 No. 52 “On the implementation of the Federal Law “On the financial recovery of agricultural producers.”

A review of these documents demonstrated the complete absence of any industry distinctions between the analyzed enterprises. Meanwhile, it should be remembered that acceptable values ​​of indicators can differ significantly not only for different industries, but also for different enterprises of the same industry, and a complete picture of the financial condition of a company can only be obtained by analyzing the entire set of financial indicators, taking into account the specifics of its activities. The approved indicator values ​​are purely informational in nature and cannot be used as a guide to action. In this regard, it is necessary to develop a regulatory framework at the level of government regulations or ministries and departments at an industry level.

Distinctive features of modern agricultural enterprises are a lack of working capital, low solvency discipline, an increase in the volume of barter transactions, and the high cost of credit resources. As a result of these and other factors, enterprises do not have the means to fulfill their payment obligations, including payment of wages, payment for goods (work, services), and debts to the budget are growing.

At the same time, even in such difficult conditions, many enterprises remain afloat. Therefore, “extreme” values ​​of indicators characterizing the structure of assets and liabilities of the balance sheet, solvency and financial stability of organizations must take into account the peculiarities of the current situation and the boundaries within which the enterprise’s management is still able to develop strategic steps to overcome the crisis without leading to bankruptcy proceedings .

The criteria in force for agricultural enterprises in the United States (since we have taken the path of borrowing the Anglo-American financial model) are also far from Russian specifics. This happens primarily for two reasons: first, the economic conditions of Russian agricultural production are very different from the economic conditions of the United States or Canada; secondly, a distinctive feature of domestic politics and agriculture is the fact that - especially among small agricultural enterprises - economic difficulties are beginning to take on a social character. Thus, the principles of a market economy are violated.

In our opinion, the main attention when adapting the traditional approach should be focused on closing existing gaps when carrying out financial analysis procedures.

The main proposals for the further development of final financial analysis procedures are as follows:

Calculation of own standards or optimal levels of financial ratios for the analyzed company using well-known methodological techniques;

Selecting a narrow (<индикаторной>) a sample of financial ratios, the composition of which may vary for different organizations;

Qualitative assessment and determination of weights of indicator indicators based on comparison with calculated optimal levels, trends, mutual comparison and accepted logical rules;

Development of a standard format for a conclusion on the financial activities of a company, which not only states the problems of the analyzed company, but also indicates the factors of current and future changes, as well as makes recommendations for overcoming, mitigating or strengthening them.

References

1. Bocharov, V.V. Financial analysis/V.V. Bocharov. - St. Petersburg: Peter, 2007. -204 p.

2. Vasilyeva, L.S. Financial analysis / L.S. Vasilyeva, M.V. Petrovskaya. - 3rd ed. - M.: KNORUS, 2008. - 816 p.

3. Efimova, O.V. Financial analysis/O.V. Efimova.-5th ed., revised. and additional - M.: Accounting, 2006.-528 p.

4. Endovitsky D.A. Diagnostic analysis of financial insolvency of organizations: textbook. allowance / D.A. Endovitsky, M.V. Shcherbakov. - M.: Economist, 2007. -287 p.

5. Methodology for calculating indicators of the financial condition of agricultural producers: approved. by Decree of the Government of the Russian Federation of January 30, 2003 No. 52-M.: Finance and Statistics, 2004.- 2 p.

6. Morozova V.L. Historical experience, or evolutionary development of economic analysis of economic activity in Russia from the standpoint of externalism / V.L. Morozova // Economic analysis: theory and practice. - 2007. - No. 16 (97). — P. 60-68.

7. Tax Code of the Russian Federation (Part 2): Chapter 26 1. Taxation system for agricultural producers (single agricultural tax) . – Legal reference system “Garant”

8. On the development of agriculture: Federal Law of the Russian Federation of December 29, 2007 No. 264-FZ

9. Savitskaya, G.V. Analysis of economic activities of agricultural enterprises: textbook. allowance/G.V. Savitskaya. — 5th ed., rev. and additional - Mn.: New knowledge, 2005.

10. Kubyshkin I. Using financial analysis for company management/ Kubyshkin I.//Financial Director. — 2005. -No. 4

11. Sokolov Ya.V. Accounting from its origins to the present day / Sokolov Y.V. — M.: Audit. UNITY. 1996.

12. Zimin N.E. Analysis and diagnostics of the financial and economic activities of the enterprise/N.E. Zimin, V.N. Solopova. M.: KolosS, 2005 -384 p.

13. Voitolovsky N.V. Economic analysis: Fundamentals of theory. Comprehensive analysis of the economic activities of an organization: Textbook / Voitolovsky N.V., Kalinina A.P., Mazurova I.I. - M.: Higher Education, 2005. - 509s

Target conducting training “Data analysis and relationship modeling in the R package” – learn the basic capabilities of the R program - a free programming language for carrying out statistical calculations, as well as learn how to organize and manage data input, conduct primary statistical analysis of data, present them graphically, and be able to find relationships in data. The training is designed for students without experience in R or with basic knowledge of the package.

It is advisable for students to have programming skills and be familiar with the basics of statistical analysis.

Upon completion of training you will be able to use the R program:

• Correctly form a sample of data for analysis
• Organize data entry and manage data
• Perform descriptive statistical analysis
• Study relationships in contingency tables
• Test statistical hypotheses about equality of means
• Use graphics capabilities
• Conduct correlation analysis
• Conduct regression analysis
• Conduct ANOVA

Duration of training: 32 academic hours. or 4 days.

Training program:

Topic 1. Basic concepts of statistical data analysis – 2 academic hours.

• Statistical research
• Methods for obtaining data
• The difference between observation and experiment
• Population and sample
• Data requirements when forming a sample
• The concept of point and interval statistical estimation
• Signs and variables
• Variable measurement scales
• Areas of statistical data analysis
• Descriptive and analytical statistics
• Selection of statistical analysis methods depending on the scales of measurement of variables
• Statistical hypothesis
• Types of statistical errors
• Principles of statistical hypothesis testing
• Choosing a significance level when testing hypotheses

Topic 2. Introduction to working in the R environment – ​​2 academic hours.

• Features of working with R
• Installing the program
• Starting the program
• R environment
• Command line and dialog interface
• Rules for specifying commands
• Creating a working directory
• Packages
• Graphical interfaces
• R as a calculator
• Help system

Topic 3. Fundamentals of programming in R – 2 academic hours.

• Types of objects in R
• Vector
• Lists
• Matrices
• Factors
• Data tables
• Expressions
• Data access operators
• Functions and Arguments
• Loops and Conditional Statements
• Database Management in R
• Vectorization of operations
• Debugging
• Object-oriented programming

Topic 4. Data entry and organization in R – 2 academic hours.

• Data download methods
• Direct data entry
• Entering data into a table
• Importing data from MS Excel
• Importing data from other statistical packages and databases
• Saving analysis results
• Specifying Quantitative Data
• Specifying ordinal and nominal data
• Setting missing values ​​in data
• Identifying outliers and errors
• Principles of data transformation

Topic 5. Graphics capabilities of R – 2 academic hours.

• Graphics functions
• Graphics devices
• Graphics options
• Interactive graphics
• Composite images
• Output devices

Topic 6. Descriptive statistical analysis in R – 4 academic hours.

• Statistics of Central Tendency
• Arithmetic mean
• Modal meaning
• Median value
• Scatter statistics
• Variance and standard deviation
• Coefficient of variation
• Percentiles
• Histograms
• Box plots
• Z-transform
• Normal distribution law
• Skewness and kurtosis
• Checking distribution for normality
• Some laws of distribution
• Binomial distribution
• Poisson distribution
• Uniform distribution
• Exponential distribution
• Lognormal distribution
• Standard error and interval for the mean

Topic 7. Formation of data for analysis using the sampling method – 2 academic hours.

• General and sample population
• Sample characteristics
• Features of the sampling method of research
• Sample classification
• Types and methods of probabilistic selection
• Sampling methods
• Simple random selection
• Systematic random selection
• Cluster selection
• Single-stage cluster selection
• Multi-stage cluster selection
• Algorithm for conducting sample surveys
• Determining the required sample size

Topic 8. Statistical tests for identifying differences in samples in R – 4 academic hours.

• Hypotheses about comparing means
• Z-test for comparison of means
• Z-test for comparison of shares
• One-sample t-test
• T-test for independent samples
• T-test for dependent samples
• Conditions for applying nonparametric tests
• One-sample Wilcoxon signed-rank test
• Mann-Whitney test
• Sign test for related samples
• Wilcoxon signed-rank test for related samples
• Nonparametric Kruskal-Wallis Analysis of Variance
• Friedman test for dependent samples

Topic 9. Assessing the relationship between variables in R – 4 academic hours.

• Analysis of the relationship between categorical variables
• Contingency tables
• Expected frequencies and residuals in contingency tables
• Chi-square test
• Agreement criterion
• Classification of types of relationships between quantitative variables
• Scatterplots
• Prerequisites and conditions for conducting correlation analysis
• Pearson correlation coefficient
• Rank correlation coefficients
• Spearman correlation coefficient
• Checking the significance of a relationship
• Interval estimates of correlation coefficients
• Partial correlation coefficients

Topic 10. Modeling the form of communication using regression analysis in R– 4 academic hours.

• Basic concepts of regression analysis
• Paired and multiple linear regression model
• Prerequisites for linear regression analysis
• Estimation of regression coefficients
• Checking the validity of the regression model
• Regression Equation Significance
• Significance of regression coefficients
• Selection of variables in regression analysis
• Assessing the accuracy of the regression equation
• Assessing the statistical stability of the regression equation
• Point and interval estimates of the dependent variable
• Nonlinear regression models
• Categorical independent variables in a regression model

Topic 11. Modeling relationships using analysis of variance in R– 4 academic hours.

• ANOVA models
• Prerequisites for using analysis of variance
• Testing the hypothesis of equality of variances
• One-way ANOVA model
• One-way ANOVA table
• Assessing the degree of influence of a factor
• Post hoc tests for paired comparisons
• Analysis of variance with two or more factors
• Two-way ANOVA table with interaction
• Graphic interpretation of the interaction of factors
• Multifactor Model Analysis

Quite often, production processes require non-measurement control. There are many reasons for this, for example, the determination of parameters that cannot be measured or the measuring equipment is too expensive, while at the same time the parameters can be easily determined by a person with a certain degree of training.

Imagine that you are buying ordinary black shoes. You decided on a model, tried on a pair - it suited you. What else will you pay attention to before purchasing?

• You'll probably want the shoes to be evenly colored;
• So that the color of the right shoe matches the color of the left;
• So that there are no rough scratches on the visible part;
• So that the stitching of the sole is smooth;

Even if we imagine that there are automatic shoe quality control devices that evaluate all of these points, most likely their cost would lead to a significant increase in the price of products. On the other hand, if the end buyer evaluates the quality of shoes “by eye,” then it would be quite logical to introduce an imitation of such control at the enterprise, and this could well replace a dozen expensive devices that evaluate the criteria listed above. In this example, an inspector assessing the quality of manufactured shoes can act as a measuring system.

But we should not forget about the training of such an employee and certain initial criteria that he must follow. Without this, each inspector will judge product quality based on personal considerations and feelings. Attributive Gage R&R makes it possible to evaluate how well the controller performs the work of the measuring system.

Attribute Gage R&R or Attribute Agreement Analysis is specifically designed to evaluate measurement systems that exclude the use of a measuring tool, when operators themselves determine the quality of products visually or by the presence of some specific criterion (absence of gross scratches on the visible part of the shoes). It goes without saying that it is impossible to assign a numerical characteristic to such measurements. Instead, the products being tested are assigned an attribute: “good” - “bad”, “pass” - “fail”, etc. This type of analysis can also be used for calibration assessment: “excellent” - “good” - “satisfactory” - “unsatisfactory” or for assessment by the presence and number of defects.

The purpose of the analysis is to assess the proximity of operators' judgments to the standard and the convergence of operators' judgments (agreement with the previous judgment). The result of the analysis will thus indicate how close the results of the measuring system are to the standard criteria and how far the operators agree with their own judgments, i.e. can the judgments of controllers be trusted?

Examples for which Gage R&R attribute analysis is applicable:

• A good example is the contact lens manufacturing process. Any optical body, one way or another, contains a certain number of aberrations, but not all of them are visible to the human eye. Each lens undergoes outgoing quality control: the operator-controller visually assesses the presence of defects/contamination using a powerful magnifying glass and special illumination. Attribute Gage R&R can be used to determine how well an examiner can distinguish lens defects from test solution contamination.
• Another example would be testing the integrity of tablets at a pharmaceutical plant. Integrity control is also carried out by a person when inspecting a blister containing up to 2000 tablets. The measurement in this case is the presence of defects or defective products in the sample.

In the example with tablets, the operator can present judgments in a binary form: “passes” - “fails,” but, due to the specifics of the sample, answers can be provided in another form. For example, up to 2 damaged tablets – “excellent”, up to 5 – “satisfactory”, up to 20 – “unsatisfactory”, above 20 – “critical”, or rate on a scale from 1 to 5.

Let's consider the analysis of attribute Gage R&R in the Minitab statistical package using the example of quality control of tablets.

The analysis assessed three operators testing 20 tablets twice. Tablet attributes were determined using standard criteria for cracks and other damage.

To carry out Attribute Agreement Analysis(attributed Gage R&R), select Stat > Quality Tools > Attribute Agreement Analysis:

Rice. 1: Attribute Agreement Analysis Configuration Window

In the program dialog box, define columns indicating operators, samples and measurement results - attributes, as shown in range 1. According to the problem, the correct attributes of the samples were known - in this case, you should enter the “Standard” column in the “Known standard attribute” field, as shown in range 2.

The analysis can be carried out without knowing the sample attributes or without repeating the experiment twice. In the first case, it will not be determined how close the operators' judgments are to the standard, and in the second, how confident the operators are in their judgments.

Click “OK” to get the analysis results:

Figure 2: Analysis results in graphical form

The Within Appraiser chart displays the percentage of operators' agreement with previous judgments. The chart will only be displayed if each operator has rated the sample two or more times, that is, if data for comparison is available. In this case, operator A is 85% confident in his judgment, although the confidence interval ranges from 60 to 97%. The higher the convergence of the operator’s judgments, the less variation he introduces into the observation results.

The Appraiser vs Standard chart shows how close operator responses are to the correct sample attributes. Even though operator A is 85% confident in his judgment, his answers are correct only 70% of the time.

In the considered example, only operator C has sufficient qualifications to control product quality: the convergence of judgments and proximity to the standard is 95%, i.e. its contribution to the variation of observations is 5%.

The analysis results in the Session window give a slightly clearer picture of how the agents did their jobs:

Rice. 3: Attribute Agreement Analysis results

Within Appraisers chart raw data from the Within Appraisers chart as well as a table of kappa coefficients. Kappa coefficients indicate how statistically confident operators are in their decisions. With kappa equal to 0, we can conclude that the operator is not confident in his judgment, his choice resembles a lottery or tossing a coin in the air. Kappa 1 means that the operator is 100% confident in his judgments; accordingly, with a coefficient of -1, the operator is completely unsure and assigns a new attribute to the sample each time. Taking into account the recommendations of the AIAG guidance on the analysis of measuring systems, a coefficient of 0.75 is sufficient to recognize the measuring system as suitable. A value of 0.4 or lower indicates that the measurement system is unsuitable.

The probability value (P-value) is estimated for the hypotheses:

H 0: operators do not agree with their decisions;
H α: Operators agree with their judgments.

Kappa and probability values ​​are fundamental when evaluating measurement systems using attribute Gage R&R. Considering the results obtained, we can conclude that based on a kappa of 0.48 and a probability value of 0.0158, the operator is confident in his decisions, but such a measuring system is not acceptable for critical control areas.

Rice. 4: Attribute Agreement Analysis results

Looking at the kappa coefficients in the Each Appraiser vs Standard analysis table, we can conclude that operators A and B have little knowledge of the evaluation criteria, but operator C has a fairly good command of product quality criteria. The Assessment Disagreement table displays points where operators' opinions differ from the standard. For example, operator A recognized 3 blisters with tablets as suitable, while the standard criteria indicate their unsuitability. Also, operator A showed different results three times on the first and second attempts, i.e. At first he determined the product to be suitable, and on the second attempt he changed his judgment. Operator B rejected quality items three times and also changed his judgment three times. Operator C changed his mind about the sample attribute only once.

The results from the Between Appraiser table illustrate how operators generally reproduce their judgments, i.e. how much they agree among themselves, without taking into account the true attributes. Agreement between operators does not at all mean that the judgments are correct.

Rice. 5: Between Appraisers and Appraiser vs Standard results tables

The All Appraisers vs Standard table, accordingly, displays how similar the judgments of all operators are to the correct attributes, i.e. the extent to which operators generally agree with the standard. As in the previous case, agreement with the standard does not mean agreement between operators.

The Kappa Statistics table indicates that in both cases the kappa coefficient is below 0.75. The testimonies of operators as a single group are statistically insignificant. Operator agreement in general with the attributes of the standard is slightly higher, but still not sufficient. The high coefficient of agreement with the standard relative to the agreement between operators is caused by the high coefficient of agreement with the standard of operator C.

Resume

• In general, the use of the analyzed measuring system is associated with a number of difficulties. Operator A makes too many mistakes, resulting in unsuitable products being considered good. At the same time, operator B evaluates products too critically, which leads to the rejection of suitable products.
• Operator C showed both the proximity of his judgments to standard ones and his confidence in them.
• From the above it follows that operators A and B need additional training or practice.

In the example considered, a very wide confidence interval is observed. which brings some uncertainty to the analysis results. To reduce the range, the study can be repeated with a larger number of samples.