# Find the moving average of the time series of quarterly production (in tons) of coffee in an Indian State as given below. After that, come...

Find the moving average of the time series of quarterly production (in tons) of coffee in an Indian State as given below. After that, come up with a trend line to approximate the production in future.
Production (in Tons)    So, T = 9.9673-0.1135 ? Quarter
number Notes:
Any number of years moving average can be used. Quarterly moving average has been chosen in this case.
Since it is not centred, centering is done as shown.
The trend can also be obtained from the time series as required here.
The summation for quarter numbers and the actual production are obtained. The additional values of
summation of x2 (quarter number squared) and summation of xy (production ? quarter number) are
obtained from the additional columns indicated.
The values of a and b of the trend line equation can then be obtained as
shown. Though not required;
It was possible to obtain deseasonalised data before obtaining the tend line. This means a better
forecasting equation is obtained (moving average and trend equation would have been used)
Seasonal factor S is obtained by averaging the error variation A/T for each quarter as per the second table.
Since the summation of the average is not equal to 4 (seasonal aspect) it has to be corrected by the factor
4/3.941.
The deseasonalised data is then obtained. Notice the way here a multiplicative model was chosen because
of the way the seasonal aspect keeps on changing.
Determination of S   raphael answered the question on January 11, 2019 at 04:50

### Related Questions

• As a tax consultant to the Government of your country, you have been asked to estimate an empirical model on the demand for income tax evasion...(Solved)

As a tax consultant to the Government of your country, you have been asked to estimate an empirical
model on the demand for income tax evasion (ITE) in the country. You think true income (TI),
Marginal tax rate (MTR), penalty rate (PR) and probability of detection (PROB) will be important
variables. Using national time-series annual data from 1967 to 1995, you estimate the following
regression equation: The calculated t – statistics are reported in parentheses, and figures are in shillings.
Required:
i) Write down the population theoretical empirical model on the demand for income tax evasion
(ITE).
ii) Provide a theoretical justification of the empirical model specification in (i) above, that is, expected
signs of the regression coefficients and why.
iii) Interpret the constant term (-52.59), coefficient for true income (33.44) and coefficient of
probability of detection (-1.48) in the context of the problem.
iv) Before collecting data, the principal tax collector and her staff believed that penalty rate had a
negative influence on income tax evasion and should therefore be used as leverage on those who
From the regression results, should this be the case? Why?
(Note: critical t-value = - 1.701 at 0.05 level of significance)

Date posted: January 10, 2019.  Answers (1)

• Briefly but clearly, explain the difference, if any, between regression analysis and correlation analysis.(Solved)

Briefly but clearly, explain the difference, if any, between regression analysis and correlation analysis.

Date posted: January 10, 2019.  Answers (1)

• Kenya Graduate School (KGS) offers a variety of graduate courses. However, its main emphasis has been on information science (IS) courses. Due to the laboratory equipment...(Solved)

Kenya Graduate School (KGS) offers a variety of graduate courses. However, its main emphasis has been on
information science (IS) courses. Due to the laboratory equipment requirements for IS courses, KGS has to
estimate in advance the expected students enrolments. Over the last 5 years, the students enrolments, by
quarter, has been: Required:
a) Determine the estimates, by quarter, for year 1996. Justify the method you use.
b) If linear multiple regression were to be used in order to determine the predicting equation, what other
variables would be included?
c) How would the expected enrolments be compared to the actual enrolments? Date posted: January 10, 2019.  Answers (1)

• In recent years environmentalists and health professionals have been concerned about the ill effects onthe environment of the widespread use of insecticides. If human beings...(Solved)

In recent years environmentalists and health professionals have been concerned about the ill effects on
the environment of the widespread use of insecticides. If human beings are to cope with the problem
and make decisions about how to deal with it, they must understand the effect of insecticides on humans
and other animals. In the kind of study often done to promote such understanding, two professors at
University of Nairobi investigated the effect of a commonly used insecticide on sheep. Among other
statistical analyses, they derived the following linear regression equation (n = 16): This equation describes the relationship between the activity of a certain enzyme in the
sheep's brain (Y)and the time (in hours) after the sheep has been exposed to the insecticide (X).
Required:
i) Suppose 30 hours have elapsed since a sheep has been exposed to the insecticide, what is the
predicted value of Y?
ii) How would you describe the relationship between the two variables? iii) The professors computed a coefficient of determination (r2) of 0.86 from the data. What
conclusion can be drawn about the true relationship between the two variables?
iv) What assumption are necessary to solve part (iii) above?

Date posted: January 1, 2019.  Answers (1)

• Does a high correlation mean that one variable causes another variable to vary?(Solved)

Does a high correlation mean that one variable causes another variable to vary?

Date posted: January 1, 2019.  Answers (1)