0% found this document useful (0 votes)
216 views22 pagesChapter 4 Functional Forms
Chapter 4 discusses various functional forms of regression models including double log, log-linear, semi-log, and polynomial regression models. It explains the significance of slope and elasticity coefficients, as well as the importance of comparing models with the same dependent variable form. The chapter also includes exercises and examples for practical understanding of these regression concepts.
Uploaded by
Khushi GargCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
216 views22 pagesChapter 4 Functional Forms
Chapter 4 discusses various functional forms of regression models including double log, log-linear, semi-log, and polynomial regression models. It explains the significance of slope and elasticity coefficients, as well as the importance of comparing models with the same dependent variable form. The chapter also includes exercises and examples for practical understanding of these regression concepts.
Uploaded by
Khushi GargCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
Chapter 4
Functional Forms of
Regression Models
After learning this chapter you will understand :
Double log Regression Model.
Comparing Linear and Log Linear Models.
Multiple Log-Linear Regression Models.
Semi-log Models
v Log-lin Model.
v¥ Lin-log Model.
Instantaneous vs Compound Rate of Growth.
Reciprocal Models.
Polynomial Regression Models.
Regression Through Origin.
Regression on Standardized Variables.
For Full Course Video Lectures of
All Subjects of Eco. (Hons), B Com (H), BBE,
MA Economics, NTA UGC NET Economics,
Indian Economic Service (IES)
Register yourself at
[Link]
Dheeraj Suri Classes
Prime Academy
9899192027 Prime Academy, [Link]
Basic Concepts
1. Functional Forms : Within the confines of the linear-in-parameter regression
models, a regression model can assume a variety of functional forms. In particular
we will discuss the following functional forms :
(i) Log-linear or Constant elasticity model,
Gi) Semi-log models,
Gii) Reciprocal models,
(iv) Polynomial regression models, and
(v)__ Regression through the origin model.
(vi) Regression on Standardized Variables.
Log-linear or Constant Elasticity Model : Consider the regression model
In¥, = 8, +B, 1nX, +u,
This is a regression model which is linear in parameters and since the variables are
in log forms, so we call it a log-linear model. Also, since both the variables are in
log forms so we call it a double log model. Here, # measures elasticity of Y with
respect to X so we also call it a constant elasticity model.
In this model the slope coefficient measures the elasticity of Y with
respect to X, ie., the percentage change in Y for a given percentage change in X.
wv
Note We cannot compare the r* of a linear model with the r* of a double
log model, because to compare the r* values of two models, the
dependent variables must be in the same form.
3. Multiple Log-Linear Regression Model : The regression model in which there
are {Wo or more explanatory variables and all variables are in log form is called
multiple log-linear regression model. For example, a three variable log-linear
model can be expressed as
In¥, = 6+ B,1nX,, + B,nX,, +4,
In this model, ; is the intercept term and /; & /; are partial slope terms. Here, >)
measures the elasticity of Y with respect to X>, keeping the effect of X3 as
constant, i¢., it measures the percentage change in Y for a percentage change in
Xz, keeping the effect of X3; as constant. Since the effect the X3 is constant,
therefore, > is called the partial elasticity. Similarly, #: measures the elasticity of
Y with respect to X3, keeping the effect of X2 as constant, i.e., it measures the
percentage change in Y for a percentage change in Xs, keeping the effect of X2 as
constant.
Exercise 1
QI. What is meant by a slope coefficient and an elasticity coefficient? What is the
relationship between the two?
Q2. Write a note on the following :
If we wish to compute elasticity directly from our model one must convert the
model to log linear or linearly log. [BBE 2011]
Econometrics 4. By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
Q3. The demand for roses (Y) depends on [Eco. (H) 2009]
X = Average wholesale price of roses.
X3 = Average wholesale price of carnations,
X: = Average weekly family disposable income.
The following demand function is to be estimated :
In¥, = B, + B, InX,, + B,InX,, + B,InX 4, +u,
Interpret the coefficients B;, B; and B, and give their economic significance.
Q4. Consider the following models :
Modell: InY; By + BilnX; +
Model 2: InY*; = Ai + AslInX*; + uw
Where, Y*; = wrYi and X*; = w»Xi, the w’s being constants.
(i) Is there a difference in the estimated intercept and slope coefficients of the
two models.
(ii) Is the r? different between the two models?
Q5. State whether the following statements are True or False. Give feasons for your
answer ; [Eco. (H) III Sem. 2013]
-¥. x, =X,-X,
(a) The regression model yy = By + Box; + ui, where Y,
must necessarily pass through origin.
(b) In the log-linear regression models, the magnitude of the estimated slope
coefficients is invariant to the units in which the explanatory variables are
measured, unlike linear models.
Q6. Following is the demand schedule for commodity x [BBE 2014]
Dy = APs, Py. Y)
Where D, is the demand for commodity x, Py is its price, P, is the price of related
commodity y and Y is the income of the consumer.
How do you measure the elasticity of demand with respect to own price and price
of related commodity Y if you use (i) double log model, (ii) linear model.
Q7. Consider the Cobb-Douglas production function: [Eco. (H) IM Sem. 2017(ER)]
O,=e°K/ Le"
Where, Q denotes output, K denotes capital input and L denotes labour input and ¢
= 2.71828.
(a) Formulate a model that can be used to estimate the parameters a, 8 and y
using ordinary least squares.
(b) Show that this model implies a constant partial elasticity of output with
respect to labour but a variable marginal effect of labour on output
Q8. State whether the following statements are True or False. Give reasons for your
answer : [Eco. (H) IV Sem 2015]
Ina double log model, the slope and elasticity coefficient are the same.
Q9. State whether the following statements are true or false. Give reasons or proofs :
Ina double log model. In Y, = A + B In X; + u, the slope coefficients are different
from elasticity coefficients. [Eco. (H) IV Sem 2016]
Econometrics 43 By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
Numerical and Conceptual Problems
Double Log Models
QI. The OLS Regression based on the log-linear data gave the following results
In, = 4.8877 + 0.1258 InX,
Se =(0.1573) (0.0148)
t =(31.0740) (8.5095)
p = (1.25x10%) (2.79x10%) r= 0.9005, n=10
where Y = Maths Score, X = Family income
(i) Interpret the intercept and slope term
(i) Interpret the coefficient of determination.
(iii) Test the significance of regression coefficients.
Q2. Based on 11 annual observations the following results were obtained :
Model A :
¥, =2.6911—0.4795X,
se= (0.1216) (0.1140) P= 0.6628
Model B:
In¥, = 0.7774 - 0.2530 In X,
se= (0.0152) (0.0494) P= 0.7448
where Y = cups of coffee consumed per person per day
X = the price of coffee in rupees per cup.
(a) Interpret the slope coefficients in two models.
(b)_ You are told that ¥ =2.43 and ¥ =1.11. At these mean values, estimate the
price elasticity for model A.
(c)__ What is the price elasticity for model B?
(d)_ From the estimated elasticities, can you say that the demand for coffee is
price inelastic?
(e) How would you interpret the intercept in model B?
(f) Since r? of Model B is larger than that of Model A, Model B is preferable to
Model A. Comment on this statement.
Q3. Consider the following models : [Eco. (H) IV Sem. 2019]
Model 1: In¥f = a + @jInX} +uy
Model Il : In¥; = By + BolnX; + 1
Where ¥/ = w,¥; and X? = wpX; , the w’s being constants.
(i) Establish the relationships between the two sets of regression coefficients
and their standard errors.
(ii) _ Is the R? different between the two models?
Multiple Log-Linear Regression Models
Q4. Using 21 annual observations, the following equation for demand for a good was
estimated using OLS :
In¥ =1.71-0.35In X, +0.47In X, R? =0.876, R
se =(0.059) (0.083) (0.083)
Econometrics 44 By Dheeraj Suri, 9899-192027
0.843, Prime Academy, [Link]
Where,
T a No. of units demanded
Xi = __ Price of goods (Rs. per unit)
X» = Consumer's income
(i) Test at @ = 5% whether the good has unit income elasticity against the
alternative that the demand for the good is income inelastic.
(ii) Test the overall significance of the regression.
Q5. The following demand equation was estimated using monthly data on mineral
water consumption, numbers in parentheses are standard errors
In, =1.534—0.750In P +0.251 In P*
se = (0.2011) (0.1012) (0.2001)
Where :
Q, = millions of one litre mineral water bottles sold
P, = price of one litre mineral water bottle
Pe = price of one litre soda beverage bottle
(i) Interpret the slope coefficients
(ii) Test at 5% level of significance, whether the demand for mineral water is
perfectly inelastic or not. (Eco. (H) III Sem. 2013]
Q6. For the data for 46 states in USA for 1992 following regression result was
obtained :
InC =4.30+1.34In P+0.171In¥
se =(0.91) (0.32) (0.20) R*
where C = Cigarette consumption packs per year
P = Real price per pack
Y = Real disposable income per capita
(i) What is the elasticity of demand of cigarettes with respect to price and
income? Are they statistically significant if not then why?
(ii) How would we obtain R? from R? given above. Then test for overall
significance of regression. [BBE 2009]
QZ. You are given the following Cobb Douglas Production function :
InY, =-1.65+ 0.34InZ, + 0.851nK,
t 2.73) (1.83) (9.06) R? =0.995 n=22
(i) Interpret the partial regression coefficients
(i) Find the returns to scale.
(ii) Test for the significance of partial regression coefficients. Will you use a one
tail or two tail test?
(iv) What can you about the overall significance of the regression model.
Q8. From the following Regression function :
In¥, =~1.5495+0.9972In X,, —0.3315In X,,
se = (0.0903) (0.0191) (0.0243) R?=0.994 n=23
where Y = final demand
Econometrics 45 By Dheeraj Suri, 9899-192027 [Link]
Real energy pric
(i) Interpret the partial regression coefficients
(ii) Test for the significance of partial regression coefficients. Will you use a one
tail or two tail test?
Gii) What can you about the overall significance of the regression model.
(i) Compute the value of adj. R? for the above model.
Q9. The demand for apple juice in USA as a function of disposable income (X2), price
of apple juice (X3), price of orange juice (X.), price of mango drink (Xs) is given
as follows for the period 1960-1982
In¥ =2.19+0.34In X, —O5lInX, +0.14InX, +0.091nX,
Rip = 0.98
It is then assumed by manufacturers that the three drinks are unrelated, i.e.,
consumption of apple juice is not affected by price of orange and mango juice :
In¥ =2.03+0.451n X, —0.381nX,
R2 =0.98
(Interpret the partial regression coefficients in the original model.
(ii). Test the manufacturer’s claim.
QIO. In the study of the demand for international reserves [i.e., foreign reserve currency
such as the US Dollar] the following regression results were obtained for 40
quarters for each of 28 less developed countries [i.e., sample size of 1120]
In(R/ P) = 0.1223 + 0.4079 In(¥ / P) + 0.5040 In o,,, — 0.0918 In oy
t = (2.5128) (17.6377) (15.2437) (-2.7449)
R? =0.8268, F=1151, n=1120
Where R = the level of nominal reserves in US Dollars
P = US implicit price deflator for GNP
YY = the nominal GNP in US dollars
the variability measure of balance of payments
the variability measure of exchange rates
(a) A priori, what are the expected signs of the various coefficients? Are the
results in accord with these expectations?
(b) | What is the interpretation of various partial slope coefficients?
(c) Test the statistical significance of each estimated partial regression
coefficient.
(@) How would you test the hypothesis that all partial slope coefficients are
simultaneously zero?
QI1. Consider the following Cobb Douglas production function estimated for Taiwan
for the period 1955-1974.
InGDP=-1 6524 + 0.3397 In L, +0.8460In K,
t = (-2.2725) (1.8295) (9.0625)
9951, RSSur = 0.0136
Where,
Econometrics 46 By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
GDP, = GDP at time t, L; = labour at time t, K; = capital at time t,
In = natural logarithms.
(i) Interpret the coefficients of the regression and comment on their individual
significance.
(i) Comment on the returns to scale experienced by the Taiwanese economy.
Gii) By imposing the restriction of constant returns to scale, the following
regression was obtained :
nf 2?) = -0.4947 + 1.0153 o( 4)
Lh L
t
t = (-4.0612) (28.1056)
R? = 0.9777, RSSr = 0.0166
Interpret the above regression.
(iv) Use a test statistic to see whether the economy is characterized by constant
returns to scale. [Eco. (H) III Sem. 2014]
QI2. Consider the Cobb-Douglas production function in its logarithmic form as follows:
In¥,= B) + BolnL; + BslnKj + Uj
where, Y = output
L = labour input
K = capital input
Suppose the following production function is estimated :
w(F) = B, +B, n( *) +y,
(i) What restriction has been imposed on the Cobb-Douglas production function
to obtain this estimated production function?
(ii) How will you test the validity of this restriction? [Eco. (H) 2009]
Q13. Consider the following model of monthly rents paid on rental units in industrial
hub cities of an economy :
In(rent) = B, + B, In(pop) + B, In(avinc) + B,(socind) +u
Where :
rent = average monthly rent paid in rupees
pop = city population
avine = average city income in rupees
socind = index of social infrastructure
(i) How will you test the hypothesis that city population and social
infrastructure have no significant joint effects on monthly rents? Explain the
steps involved to the above model.
(ii) Suppose b, is estimated as 0.066. What is wrong with the statement : “A
10% increase in population is associated with a 6.6% increase in monthly
rent.” [Eco. (H) II Sem. 2013]
QI4. Using data on labour and capital for 123 firms, the Cobb-Douglas production
function is estimated and following regression results are obtained (standard errors
are given in parentheses) : [Eco. (H) IV Sem. 2016]
Econometrics 47 By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
LN_OUTPUT; = 0.790 + 0.988 LN_CAPITAL; + 0.204 LN_LABOUR;
(se) = (0.008) (0.325) (0.1008)
R°=0.979
Where
LN_OUTPUT = log of output,
LN_LABOUR = log of labour,
LN_CAPITAL = log of capital
(a) Name the functional form of the regression model.
(b) _ Interpret the slope coefficients.
(c) What is the nature of returns to scale exhibited ?
(@) Perform an appropriate test (at 10% level of significance) to check if slope
coefficient on LN_LABOUR is significantly greater than 0.2.
QIS5. The following regression model was estimated using annual time-series data for
the period 1990-2012 for a certain country : [Eco. (H) IV Sem. 2017]
In¥, = by + bp InXp¢ + byInX 3,
Where Y, = demand for cheese (in kg.)
posable income (in Rs. (000)
X3 = price of cheese (in Rs. per kg.)
The results are summarized in the following table :
Coefficient Standard error
Intercept 2.03 0.116
Xr 0.45 0.025,
X3 —0.377 0.063
(i) Interpret the partial slope coefficients.
(ii) __ If the calculated F statistic for the estimated model is 492.513, what is its
R»
QI6. Consider the following regression model : [Eco. (H) IV Sem. 2017]
InY = Bo + By In(X:) + Bo [n(X2) + Bs In(Xs) + Bs In(X4) + ui
Where
Yc consumption of potatoes in kg.
X income in Rs. ‘000,
X2 = price of potatoes in Rs. per kg.
X; = price of cauliflower in Rs. per kg.
X,= price of cabbage in Rs. per kg.
(i) — How will you test the joint hypothesis that potato consumption is not
affected by the prices of cabbage and cauliflower ? Explain the steps
involved in the test with reference to the above model.
(ii) If the estimated value of b; is 200, it means "a 1% increase in income is
associated with a 200% increase in per capita consumption of potatoes,
everything else kept constant.” Is the above interpretation correct ? Explain.
Econometrics 48 By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
Basic Concepts
1. Semi Log Models : The semi-log models are used to measure the growth rate.
There are two forms in semi-log model.
(@ —_Log-lin Model, and (ii) Lin-log Model.
2. Log-lin Model : Log-lin model is used to measure the growth rates. Let us
consider the compound interest formula :
¥,=¥(1+r)'
Where, ¥; is the value of Y at time t,
Yo is the initial value of Y
ris the compound growth rate of Y.
Now, we can convert this equation into a regression model by taking the log on
both the sides :
In¥, =InY, +[Link]+r)
Now, let
Pr=In¥o and f=In(1 +r)
So, the regression model takes the form :
Iny, = 8+ Bt
Now, if we add the error term, we obtain the regression model as :
In¥, = B+ Bt+u,
‘This regression model is linear in parameters and it is called log-lin model because
the dependent variable Y is in log form and the independent variable ¢ is in linear
form. In this model the slope term /; measures the proportional or relative change in
Y for a given absolute change in the explanatory variable, time in the present case.
Instantaneous Versus Compound Rate of growth : The slope term / in the
w
regression model, In¥, = 8, + B,4+u,, measures the instantaneous growth rate
whereas rin the equation #; = In(1 + r) represents the compound growth rate.
4. Lin-log Model : The regression model which is of the following form :
¥, =f. +.nX, +u,
is called lin-log model
This regression model is linear in parameters and it is called lin-log model bec:
the dependent variable ¥ is in linear form and the independent variable X
logarithmic form. In this model the slope term / measures the absolute change in
Y for a percentage change in the given explanatory variable X.
Exercise 2
Theory Questions
QI. Explain the log-lin and lin-log models with examples. [BBE 2011]
Q2. What is instantaneous growth rate?
Q3. What is compound growth rate?
Econometrics 49 By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
Q4. The Reserve Bank of India continuously monitors the rate of growth of money supply
in the Indian Economy. In this context, explain the use of semi-log growth models in
estimating instantaneous and compound rate of growth of money supply. What
purpose would a linear trend model serve in this context? [Eco. (H) III Sem. 2013]
Q5. Write short note on : Log-Lin versus Lin-Log regression models. [BBE 2014]
Numerical and Conceptual Problems
QI. From the data based on population of USA (million of people) for the years 1975
to 2007 the following regression model was obtained :
In) = 5.3593+ 0.0107
t = (3321.13) (129.779) r= 0.9982
Where Y = Population of USA (millions of people)
t = time period (in years)
(i) Interpret the Intercept term (ii) Interpret the Slope term.
Gi) Find the instantaneous growth rate as well as the compound growth rate.
(iv) _ Test the significance of regression coefficients.
(vy) Interpret P.
Q2. Regressing population growth on time variables we get the following results for
the period 1980-2000 : [BBE 2013]
Incpop) = 5.3170 + 0.00743t
= (8739.39) (285.9) = 0.9996
w Identify the model.
(ii) Calculate the compound rate of growth over 21 period and interpret it, where
Antilog (0.00743) = 1.00746
(iii) Calculate the instantaneous growth rate
Q3. Consider the following equation :
In(Sal), 5.10+0.100ED, +0.110EXR
se (0.025) 0.050)
2=0.48, n=28
Where In(Sal),
ED, = Years of education of i®* worker
EXP, = Years of experience of i* worker
(i) Interpret the equation. Make appropriate hypothesis for signs of coefficients
and test your hypothesis.
(i) What are the elasticities of salary with respect to education and experience?
(iii) If we run a linear regression instead of log-linear regression then how would
the interpretation change. [BBE 2009]
Q4. To determine how expenditure on services (Y) behaves if total personal
expenditure (X) rises by a certain percentage, the following regression model was
obtained :
12564.8+1844.22 In X,
se= (916.351) (114.32) P= 0.881
Econometrics 410 By Dheeraj Suri, 9899-192027
log of salary of i worker Prime Academy, [Link]
(i) Interpret the Intercept term
Gi) Interpret the Slope term.
(iii) Test the significance of regression coefficients.
(iv) Interpret r°.
Q5. Consider the following regression results for cross sectional data for 55 rural
households in India. The regressand in this equation is expenditure on food and the
regressor is total expenditure (a proxy for income) [Eco. (H) 2010]
FEXP = -1283.912+257.27|n(fEXP)
t = (-4.3848)* (5.6625) r= 0.3769
Note : *denotes an extremely small p-value.
(What is the interpretation of coefficient of In (TEXP)?
Gi) Would you say that Engel’s Law is validated for this sample? Explain
Q6. Consider the following population regression function
In(Div), = B, + By n(PRFT), + B,Time+u,
Here, Div = Corporate Dividends Paid
PRET = Corporate Profits
In = Natural Logarithms
The estimated sample regression results for an economy for 244 quarterly
observations are presented below :
Coeff. Standard t-statistic | Prob-value
Errors
Intercept 0.4357 0.1921 2.2674 0.0243
InPRFT) 0.4245 0.0777_— 5.4614 (0,000
Time 0.0126 = 0.0014 8.93 0.0000
R?=0.9914, adj. R? = 0.9913,
Sum of Squared Residuals = 4.2657, F-Statistic = 13930.73
SE of Regression = 0.133 Prob(F-Statistic) = 0.0000
Durbin-Watson statistic = 0.0201
(®) — Whatare the economic interpretations of 2, and #,?
(i) On what counts would a researcher be satisfied with these results at a first
glance? Verify your conjectures using formal tests, For tables take the
closest value of n. [Eco. (H) 2010]
Q7. Based on the following data :
y | 86 | 79 | 76 [69 | 65 [| 62 | 52 [ si [ 51 | 48
x [3 7 [12 [17 [25 [35 | 45 [55 [70 | 120
(Estimate the model :
tea + BAX, +1,
y
(i) Interpret Bs.
(iii) Find the rate of change of Y with respect to X.
(iv) Find the elasticity of Y with respect to X at mean values of Y and X.
(v) For the same data run the regression
Econometrics 411 By Dheeraj Suri, 9899-192027 Prime cademy, [Link]
7)
¥,=B,+B,|—|+u,
“LX,
(vi) Can you compare the 17's of the two models.
[Ans. : (i) ; =0.0130-+ 0,0000833x,, (ii) B
aX (B,+B,X,)"
(iv) -0.1915, — (v) ¥ =ssas71 112.7972, (vi) No]
Q8. Based on the data on GNP and money supply for the period 1965-2006 for India,
the following regression results were obtained by regressing GNP (in billions of
Rupees) on money supply (in billions of Rupees) for alternate models
Model Intercept Slope Coefficient R?
Coefficient (f,) &,
Log-linear 0.8726 0.7839 0.927
(1.40) (108.93)
Log-lin 6,2392 0.0002 0.852
(75.85) (12.07)
Lin-log 14299 2383.4 0.879
(14.45) (16.84)
Linear 603.28 0.3718 0.921
(7.04) (55.58)
Where the figures in parentheses are ratios
(i) For each model. Interpret the slope coefficient.
(ii) For each model, estimate the elasticity of GNP with respect to money supply
(sample means of the GNP and money supply are 5113.65 and 9347.53
respectively.
(iii) Are all R? values comparable? If not, which ones are? [Eco. (H) IV Sem. 2015]
Q9. A study is undertaken to find the impact of average income of the houscholds
(INCOME, in rupees) and price of LCD television (PRICE, in rupees) on
household demand for LCD televisions (DEMAND, in *000 units) for 23 states of
India, The OLS regression gives the following results, with p-values given in
paranthesis and LN denotes natural log : [Eco. (H) IV Sem. 2017(ER)]
DEMAND, = 387.56 ~58.973LN _ PRICE, + 64.125LN _ INCOME,
() (0.015) (0.007) (0.0017)
R? = 0.6478
(a) What is the (partial) impact of price and income on demand for LCD
televisions?
(b) Test the overall significance of the model at 5% level of significance. State
the null and alternative hypothesis clearly.
QUO. The following are the regression results of foreign direct investment (FDI, in lakhs
of rupees) over time (7) for the yeas 1991. 2015 (standard errors are mentioned in
parentheses) [Eco. (H) IV Sem. 2016]
Econometrics 412 By Dheeraj Suri, 9899-192027 (se) = (0.002) (0.0001)
(a) Compute the instantaneous rate of growth in FDI.
(b) Can compound rate of growth be computed ? If yes, how ?
QIL. Using annual time-series data for the company 'Pure Juice’ for the period 2000 —
2016, the following equation was obtained : [Eco. (H) IV Sem. 2018]
In¥, = 1.2028 + 0.0214t
Se = (0.0233) (0.0025)
Where Y, = revenue of the company in * crores at time f and In indicates natural
log.
(Interpret the estimated coefficients.
Gi) Explain how the annual compound growth rate in revenues of the company
during the period can be obtained ?
(iii) Using the estimated model, how can the forecast revenue for the year 2017
be obtained ?
QI2. The sales manager of a company believes that the district sales (S,) of motor
vehicles has been growing according to the model S, = So(1 + g)!, where f is the
time. Average sales is 50 units and average time is 4 years. He obtains the
following OLS regression results [Eco. (H) IV Sem. 2021]
InS, = 3.6889 + 0.0583t
(i) What is the estimate of the instantaneous and compound growth rate?
(ii) What is the estimate of So?
(iii) What will be the elasticity of sales with respect to time?
(iv) Suppose the researcher modifies the above equation and estimates the
following regression: 5; = 5.6731 + 2.7530¢ Interpret the model.
(v) Compute elasticity of sales with respect to time for the model in part iv.
Compare your results with the answer obtained in part iii.
Basic Concepts
1 Reciprocal Models : The regression model which is of the form +
¥,=B, {Lon
x;
is called reciprocal model. This model is non linear in X because it enters the
model in reciprocal form but it is a linear regression model because the parameters
are linear.
The reciprocal model has these features : As X increases indefinitely, the
term af) approaches zero and Y approaches the limiting or asymptotic value
Bj. Therefore, reciprocal models have built in them an asymptote or limit value
that the dependent variable will take when the value of the X variable increases
indefinitely.
Econometrics 413. By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
(1
The slope of reciprocal model ¥; = By +B +4; is:
ay 1)
a a. implying that if Bz is positive, the slope is negative throughout,
and if Bz is negative, the slope is positive throughout.
Exercise 3
Theory Questions
QI. Write a note on reciprocal Models. [BBE 2009]
Q2. Discuss the reciprocal Models. [BBE 2011]
Q3. Consider the following model [Eco. (H) 2009]
1
y; = Bi +B, (Z) ui
(i) Whatis the asymptotic value of the dependent variable?
Gi) Draw rough sketches of the curve given, of the following information
(a) Bi <0,B).>0 (b) By >0,B. <0
Q4. Find the slope and elasticity of Y with respect to X for the following functional
forms: [Eco. (H) III Sem. 2012]
(a) InY =B,-B.(1/X)
(b) Y=B) +BsInX
Q5. Consider the following functional form :
¥ = By + BX +B, (4)
(i) Derive the expression for the marginal effect of ¥ with respect to X.
(ii) Derive the expression for elasticity of Y with respect to X and express it in
terms of X only.
(iii) _ Assume without loss of generality, By = 0 and By > 0, Bs > 0, For what value
of X will this function attain a minima? Draw a rough sketch for the
function. (Eco. (H) 2012]
Q6. For the log-inverse and lin-log functional forms given below, find the slope and
elasticity of ¥ with respect to X : [Eco. (H) IV Sem. 2017]
@ ~~ in(Y)=B, +B, (1/X)
(ii) Y=B) +B, In(X),
Numerical and Conceptual Problems
Ql. Yi= Bi + BoC1/Xi) X, Y non-zero
(i) Is this a linear regression model?
(ii) What is the behavior of Y as X tends to infinity?
(iii) What are the various possible cases in this model? [BBE 2013]
Q2. Given the following model [BBE 2014]
CM; = 81.79436 + 27273.17(1/PCGNP)
SE =(10.8321) (3759.99)
P= 0.4590
Econometrics 414 By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
cM = child mortality rate, i.e., number of deaths per “000 live births.
PGCNP = per capita GNP in rupees.
(i) What kind of a model is it?
(ii) Interpret the model.
(iii) At what level of PCGNP will a country achieve a child mortality value of
zero?
Q3. Based on annual percentage change in wage rates, Y and the unemployment rate, X
for united kingdom for the period 1950-1966 the following results were obtained :
¥, =~1.4282 48.2743
x;
se=(2.0675) (2.8478)
Interpret the asymptotic intercept value and explain its economic significance.
[Eco. (H) 2009]
Q4. Based on annual percentage change in wages (Y) and percent annual
unemployment rate, X for the years 1950 to 1966, following regression was
obtained :
—1.4282 + 8.2743 i
x
se= (2.0675) (2.8478)
P = 0.3849, FL, 15) =9.39
(i) Interpret the above regression.
(i) What would be the slope of the regression? What would be some likely
shapes of the curve corresponding to the above regression?
(iii) What is the elasticity of Y with respect to X at mean values of ¥ = 4.8% and
X=15%. [Eco. (H) III Sem. 2014]
Q5. Based on annual percentage change in wage rates, Y and the unemployment rate, X
for united kingdom for the period 1950-1966 the following results were obtained :
¥, =~1.4282+8.2743
x;
se = (2.0675) (2.8478) 1° = 0.3849, F=939
(i) What is the interpretation of 8.2743.
(ii) Test the hypothesis that the estimated slope coefficient is not different from
zero. Which test will you use?
(iii) How would you use the F test to t
(iv) Given that ¥ =4.8 percent and
of Y at these mean values?
(v) What is the elasticity of Y with respect to X at these mean values.
(vi) How would you test the hypothesis that true r= 0?
Q6. Change in inflation rate (m: - m1) between the two time periods is regressed on
current unemployment rate (UN,) and the following regression was obtained
the preceding hypothes
=1.5 percent, what is the rate of change
Econometrics 415 By Dheeraj Suri, 9899-192027 1 — =(-2.9715) (3.0625) 2067
(i) Comment on the statistical and economic significance of the equation.
Gi) Calculate the natural rate of unemployment.
(iii) What would be the change in inflation if the unemployment rate increases
indefinitely? [BBE 2011]
Q7. The percentage change in the index of hourly earnings (Y) and the civilian
unemployment rate (X) for the United States for the years 1958 to 1969 gives the
following regression model :
¥, =—0.2594 + 20.5880.
1
t= (0.2572) (4.3996) P= 0.6594
(i) Whatis the wage floor.
(ii) Interpret the Slope term.
(iii) _ Test the significance of regression coefficients.
(iv) Interpret
(v) The linear model for the same data is
Y =8.0147 - 0.7883X,
= (6.4625) —_(-3.2605) P=05153
(a) _ Is positive slope in the reciprocal modal analogous to negative slope
in the reciprocal modal.
(b) Compare the slope terms of two models.
(©) Compare r° for two models.
Q8. Two models for Engel expenditure function are estimated.
ModelT: — ¥, = 1087.930 + 0.077X,
t = (25.58) (21.64) R?= 0.350 F = 468,645
Model If: ¥, = 4005.077 + 0.3381/X,
t = (19.259) (20.816) R?= 0.333 F=433,310
where Y; = expenditure on food in rupees [Eco. (H) IV Sem. 2019]
X, = expenditure in rupees
(i) Interpret all coefficient value of the two models.
(ii) Are the sign of the coefficients in the two models contradictory?
(iii) Can we compare the results of the two models?
(iv) Diagrammatically show the sample regression function in the above model.
Basic Concepts
1. Polynomial Regression : The term “linear” in linear regression means that the
regression function is linear in the coefficients #1, f2, ... Bp. It is not required that
the X; appear as linear terms in the regression function. For example, we may
include power transforms of the form X‘for integer values of g. This allows us to
fit regression functions that are polynomials in the covariates
Econometrics 416 By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
For example, we could fit the following cubic model
E(Y | X) =A + 2X + BX? + fax?
This is a bivariate model, as Y and X are the only measured quantities. But we
must use multiple regression to fit the model since X occurs under several power
transforms.
The general k" degree polynomial regression may be written as
Y, = By +B,X, +B,X?+B,X} +....4B,X}
Estimating the Total Cost Function : Let Y stands for total cost and X stands for
the output, mathematically, the total cost function can be expressed by the
following cubic polynomial :
Y, =B,+B,X,+B,X?+B,X;
As a matter of fact, it can be shown that the parameters of above cubic cost
function must satisfy the following restrictions if one is to observe the typical U-
shaped short-run marginal and average cost curves :
@) Bi, Brand By each is greater than zero,
Gi) Bs <0, and
(iii) _B? <3B,B,
Theory Questions
QI. Explain Polynomial regression.
Numerical and Conceptual Problems
QI. The following regression considers the relationship between lung cancer and
smoking for 43 states in India :
¥, = B+ BX, + BX; +u,
Where, Y = number of deaths from lung cancer.
X = number of cigarettes smoked.
Results are as follows
Predictor Coeff. Std. error t p
Constant -6.910 6.193 1.12 0.271
x 1.5765 0.4560, 3.46 0.001
x? -0.019 0.008 2.35 0.024
? = 0.564, R? =0.543
F D
Residual sum of squares 311.69 26.56 0.00
Sum of squares regression 403.89
(Interpret the above regression.
(ii) Test the individual significance of regression coefficients. Which test do you
use and why? (use ot = 5%)
(iii) Construct an ANOVA table for the problem and test for the overall
significance of the model. (use a = 5%) (Eco. (H) 2011]
Econometrics 417 By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
In order to test whether the developing economies are catching up with the
advanced economies or not, a researcher regressed the growth rate of GDP of a
country on its relative per capita GDP for 119 developing countries. The relative
per capita GDP of a country is measured as a ratio of the country’s per capita GDP
to the GDP per capita of USA. The regression results were obtained as under
(standard errors are reported in parentheses) : [Eco. (H) 2013]
G = 0.013 + 0.062P, — 0.061P?
se.=(0.04) (0.02) (0.033)
R? = 0.053, adjusted R? = 0.036
Where, G is the growth rate of GDP (in %)
And, P is the relative per capita GDP (in %)
(i) Interpret the above regression results.
(ii) Find the marginal effect of P on G.
Q3. The OLS regression results based on the Cost (Y) and Output (X) are as follows :
¥, =141.7667 + 63.4776X, -12.9615X? +0.9396X;
se= (6.3753) (4.7786) (0.9857)_—_(0.0591)
R?=0,9983, n= 10
(i) Does this model represent the cost function, explain by testing the
coefficients in the model.
(ii) Test the overall significance of the regression coefficients.
(ii) Construct an ANOVA table for the problem and test for the overall
significance of the model. (use a = 5%)
(iv) Find the average and marginal cost curves.
Q4. The relationship between gross domestic product (GDP) in lakhs of rupees and the
expenditure on infrastructure (INFRA) in lakhs of rupees for 40 countries for the
fiscal year 2011-12 is postulated by the following two alternate models :
Model A: GDP =a, + @,INFRA +u,
Model B : GDP = fi, + 8, INFRA, + Bz INFRA’ +v,
If R* (Model A) = 0.5823 and R? (Model B) = 0.8823, which model would you
choose — restricted or unrestricted ? Use 5% significance level. State the null and
alternative hypothesis clearly. [Eco. (H) IV Sem. 2017(ER)]
QS. Consider the model [Eco. (H) IV Sem. 2022]
Y; = Bi + BrXoi + BsXai + Uy
Where,
Y, is the long term consumption measured in Rs. Thousand
Xj is the income measured in Rs. Thousand
X;\ is the age measured in years
(i) How will the estimated intercept and slope coefficients change if the unit of
measurement of income is changed to Rs. Lakhs.
Econometrics 418 By Dheeraj Suri, 9899-192027 Prime Academy, [Link]
(ii) Suppose the researcher thinks that usually consumption increases with
income but at a decreasing rate and consumption increases with age. How
would he modify the model to see whether the data supports his hypothesis?
(iii) Suppose the researcher wants to assess the relative importance of age and
income on Iong term consumption, what model should he estimate? Explain.
Basic Concepts
1. Regression Through Origin : There are occasions when the two-variable PRF
assumes the following form :
Y= B,X,+u,
In this model the intercept term is absent or zero, hence the name regression
through the origin.
‘The SRF of regression through origin can be written as :
¥,=2,X, +i,
On applying the OLS method to the above equation, we obtain the following
formulas for , and its variance
my XY, F a
> varlZ,J=
ee ee
2. for Regression-through-Origin Model : The conventional 1? is not appropriate
for regressions that do not contain the intercept. But one can compute what is
known as the raw r? for such models, which is defined as :
xy,
x
Although this raw r* satisfies the relation 0
