Анализ экономических факторов в регрессионной модели, влияющих на чистый доход ПАО банк ВТБ

In recent years many companies struggled to stay profitable especially in the tough economic conditions, such as in Russia. After economic crisis of 2008 most banks met a serious problems with remaining on their top positions in the ranking, VTB is not exception, their net profit fell to negative figures for the first time of their history of annual reporting in 2009. That’s why it is rational to estimate the environment indicators such as economic or financial ones in order to understand which factors influence companies, in the case of this work a commercial bank. By analyzing historical events and figures, future values can be predicted as well as estimated the relevance of the variables. A lot of factors can influence banking activity but in the scope of this work were chosen 5 economic factors, obviously 5 factors are not the maximum amount, that’s why this model can be further developed and modernized.

Indicators estimated in the scope of the report are presented in the Table 1.

Table 1 – indicators definition

Y	VTB NET PROFIT
X1	Lending interest rate (%) in Russia
X2	Foreign direct investment, net inflows (BoP, current US$) in Russia
X3	Unemployment, total (% of total labor force) (national estimate) in Russia
X4	Official exchange rate (LCU per US$, period average) in Russia
X5	GDP per capita (current US$) in Russia

As for the VTB net profit data there was met an obstacle of nonexistence of the financial reporting for this company before 2003, which made the amount of observations small and inaccurate, but this model can be updated in further years. Banks operations are highly dependable on the loans revenue that’s why there is a prediction that lending interest rate can influence banks profit. Banking sector is the part of the economy that’s why GDP per capita and unemployment rate with exchange rate are able to influence its productivity. One more important factor is foreign direct investments which can influence the company`s development, as well as cash inflows in the market. Applying econometric model with the regression equation these indicators were analyzed all provided data is annual from 2003-2016.

Indicators of the regression equation can be estimated by the method of least squares. The essence of this method lies in the finding coefficients of the model (с i ), under which the sum of the squared deviations is minimized of empirical (real) values of the influencing factors from theoretical ones, found by the determined equation of regression. [1]

Regression analysis was done via the Eviews 9.0 software.

The equation of VTB net pfofit:

Y1 = 141.441836833- 30.2416211352*X1 +2.20847048056e-09*X2

+4.16844182487*X3 +7.31091336723*X4 -0.0103887587654*X5

The figures are representing the coefficients of the model, some of the variables such as «X» will be excluded due to irrelevance of the variable on the described Y.

Table 2 regression analysis

System: VTB

Estim atio n Method: Least Squares

Date: 03/02/17 Time: 02:01

Sample: 2003 2014

Included observations: 12

Total system (balanced) observations 12

	Coefficient	Std. Error	t-Statistic	Prob.
C(1)	141.4418	108.5135	1.303449	0.2402
C(2)	-30.24162	4.428605	-6.828702	0.0005
C(3)	2.21E-09	9.01E-10	2.451586	0.0497
C(4)	4.168442	8.396837	0.496430	0.6373
C(5)	7.310913	4.412233	1.656964	0.1486
C(6)	-0.010389	0.005244	-1.981087	0.0949
Determinant residual covariance		115.5728

Equation: Y1=C(1)+C(2)*X1+C(3)*X2+C(4)*X3+C(5)*X4+C(6)*X5

Observations: 12

R-squared Adjusted R-squared S.E. of regression Durbin-Watson stat

0.947167  Mean dependent var    41.44167 0.903140  S.D. dependent var     48.85058 15.20347  Sum squared resid      1386.874 2.636985

Based on regression analysis have been identified, that R – squared 94.71% (demonstrates high tightness of the link between Y and X). Probability of variables (Prob.) less than 5% which means high level of significance for some variables, in order to estimate relevance t-statistic is used in which The significance of the variable can be tested with application of the Student criteria «t кр » with lies in proving the equation kl^^ . The t-criteria was found by the formula «=TINV(0,1;12-5-1)», where 0,1 is the probability, 12 is the number of estimation, 5 amount of regressors. A T criterion equals to 1,943180281. Under t-statistics test only c(1), c(4),c (5) have passed the test, while under testing of standard deviation to coefficients, the values of c(1), c(2), c(5), c (6) showed the level of signifisence. Applying both test it can be stated that c(1) and c(5) coefficients and variables are significant to the model and thus equation of the VTB will be equal:

Y1= 141.441836833 - 0.0103887587654*X5

And this variable is the GDP per capita, which means that this bank is highly influenced by the economic situation in Russia.

Durbin – Watson test equals = 2,64, this value shows that found model is in the area of uncertainty, it means that some of the variables may reflects on final results (correlation).

To accept model should be analyzed test on heteroscedasticity, it means that variance of the random error should constant or focused. If errors distributed without similarity it means that we can’t accept model. But with the obstacle of the lack of data on VTB financial existence the amount of observations is insufficient to determine homoscedasticity.

To sum up, the model showed that there is a relation between economic factors on the profit of the commercial bank but due to the short existence of it there yet not sufficient data for further testing, this model could be developed to estimate relevance of the economic sector on the financial results of the firms.