Эконометрика. Практическая РИУ, МНЭПУ, МЭИ

Practical exercises on the subject of "Econometrics" code (EN)

Task 1. Enterprises region (number of the enterprise A) are ordered in terms of output. Indices in the number of administrative staff is characterized. The data are summarized in Table. According to the table are calculated using the least squares linear regression coefficients.
Table.

X 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
In 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3

X 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Y 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5

Task 2. Calculate What is the sum of the squares, explaining the model ESS, if the total sum of squares of TSS = 0.204705, and the residual sum of squares RSS = 0.161231?

Task 3. Data Targets 1, calculate the correlation coefficient.

Task 4. We were rated change of the dependent variable (assuming costs) of the independent variables (income and prices) in the form of:

It can be interpreted as the coefficients of the independent variables?

Task 5. Gaussian distribution is symmetric with respect to zero, and it suggests that positive errors as likely as negative; At the same time, small errors are more common than big ones. If the random error has a Gaussian distribution with a parameter, with its value likely to be concluded between and. At what intervals will be located a random error for the same value of probability, if,?

Task 6. When and on what basis we can talk (argue) a preference for one-sided test, compared with two-way when used as an alternative hypothesis?

Task 7. data on the size of aggregate disposable income and total expenditure on private consumption in the United States between 1970 and 1979 (in billions of. Dollars in 1972 prices), the estimated linear relationship model is given.
Let us imagine that we are in 1979 and we expect to increase in 1980, the total disposable income (the same price) to billion. Dollars. Then the model projected on the selected volume of the total private consumption expenditure in 1980 is so that if you choose the level of trust,
What will be equal, and the confidence interval for the specified value?

Task 8. Consider the three options the forecast of consumption (y) appliances of revenue (x). We have:
• observed values;
• values \u200b\u200bobtained for model built without autocorrelation of errors;
• the value obtained by the model parameters are adjusted to account for autocorrelation of errors;
• values \u200b\u200bwhich alternative models for the prediction should be preferred if the mean square differences by using the three methods of calculating the values. These mean squares are equal to?

Task 9. The administration of the insurance company decided to introduce a new type of services - insurance in case of fire. In order to determine tariffs based on a sample of 10 cases of fires analyzed the dependence of the value of the damage caused by a fire on the distance to the nearest fire station:

№ p / p 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
The total damage million. 26.2 17.8 31.3 23.1 27.5 36.0 14.1 22.3 19.6 31.3
The distance to the nearest station, km 3.4 1.8 4.6 2.3 3.1 5.5 0.7 3.0 2.6 4.3
9.1. Construct a scatterplot result value (total damage), and the independent variable (the distance to the nearest station)
9.2. Determine the parameters a and b of the equation pair linear regression.
9.3. Calculate the linear correlation coefficient.
9.4. Check the statistical significance of the regression coefficient «b» using the Student t-test
9.5. Rate the statistical significance of the regression model was built with the help of the whole F-Fisher criterion

Task 10. The available data on prices and dividends on ordinary shares, also on the profitability of the company.
Number stock price (US $)% return on equity level of the dividend%
January 25 15.2 2.6
February 20 13.9 2.1
March 15 15.8 1.5
4