Hayley Brown

Applied Spatial Econometrics

Fall 2014

Lecture 3 Assignment
Due: 03 Oct 2014 @ 11:59PM

Uif!gjmf!DFPTBM3/Sebub!dpnubjnt!ebub!pn!288!dijfg!fyfdvujwf!pggjdfst!bne!dbn!cf!vtfe!up!fybnjnf!
uif!fggfdut!pg!gjsn!qfsgpsnbndf!pn!DFP!tbmbsz/!!
!
b/ Ftujnbuf!b!npefm!sfmbujnh!bnnvbm!tbmbsz!up!gjsn!tbmft!bne!nbslfu!wbmvf/!Nblf!uif!
npefm!pg!uif!dpntubnu!fmbtujdjuz!wbsjfuz!gps!cpui!jnefqfnefnu!wbsjbcmft/!Wsjuf!uif!
sftvmut!pvu!jn!frvbujpn!gpsn/!!
A model of the constant elasticity variety is linear in elasticities. Elasticities are percentage
changes. So a constant elasticity model would be:
log(salary) = 0 + 1log(sales) + 2log(mktval) + u.
In this case, the constant elasticity equation is:
log () = 4.62 + 0.162log(sales) + 0.107log(mktval)
n = 177
R2 = 0.299

!
c/ Bee!qspgjut!up!uif!npefm!gspn!qbsu!b/!Wiz!dbn!uijt!wbsjbcmf!npu!cf!jndmvefe!jn!
mphbsjuinjd!gpsn@!Wpvme!zpv!tbz!uibu!uif!qfsgpsnbndf!wbsjbcmft!pg!uiftf!gjsnt!fyqmbjn!
nptu!pg!uif!wbsjbujpn!jn!DFP!tbmbsjft@!!
This variable cannot be included in logarithmic form because nine of the companies in the
sample have negative profit values. It is impossible to take the log of a negative value, as it is
undefined. The new model is as follows:
log () = 4.69 + 0.161log(sales) + 0.098log(mktval) + 0.000036(profits)
n=177
R2 = 0.299
Given the value of R2, I surmise that together these variables account for nearly 30% of the
sample variation in log(salary). This is certainly not most of the variation. Moreover,
profits seem to add very little to the model, suggesting that profits exert negligible influence
over log(salary).

!!
d/ Bee!uif!wbsjbcmf!dfpufn!up!uif!npefm!jn!qbsu!c/!Wibu!jt!uif!ftujnbufe!qfsdfnubhf!
jndsfbtf!jn!tbmbsz!gps!bnpuifs!zfbs!pg!DFP!ufnvsf-!ipmejnh!puifs!gbdupst!gjyfe@!!
log () = 4.56 + 0.162log(sales) + 0.102(mktval) + 0.000029(profits) + 0.012(ceoten)
n = 177
R2 = 0.318

The implication is that when CEO tenure increases by one year, salary increases by 1.2%.

e/ Gjne!uif!tbnqmf!dpssfmbujpn!dpfggjdjfnu!cfuxffn!uif!wbsjbcmft!mph)nluwbm*!bne!qspgjut/!
Bsf!uiftf!wbsjbcmft!ijhimz!dpssfmbufe@!Wibu!epft!uijt!tbz!bcpvu!uif!PMT!ftujnbupst@!!
The sample correlation coefficient is 0.78. This suggests that log(mktval) and profits are
highly correlated. Give this close correlation, it is difficult to estimate the independent effect
of each on log(salary). This does not indicate bias in the OLS estimators, though it may cause
their variances to be large. It is worth noting that profit is a short-term measure, while mktval
is based on past and current profitability; however, mktval also incorporates future
expectations of profitability, which may or may not come to fruition and should scrutinized in
their own right.
Code
#**********a**********#
# Load data
> load("/Users/Abulafia/Documents/school 2/Fall 2014/Applied Spatial
Econometrics/ceosal2.RData")

# Rename things to make them more accessible
> ceosal2 <- data
> View(ceosal2)
> salary <- ceosal2$salary
> sales <- ceosal2$sales
> mktval <- ceosal2$mktval
> lsalary <- log(salary)
> lsales <- log(sales)
> lmktval <- log(mktval)


# Fit the model
> fit <- lm(lsalary ~ lsales + lmktval, data=ceosal2)

# Obtain values for equation
> summary(fit)

Call:
lm(formula = lsalary ~ lsales + lmktval, data = ceosal2)

Residuals:
Min 1Q Median 3Q Max
-2.28060 -0.31137 -0.01269 0.30645 1.91210

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 4.62092 0.25441 18.163 < 2e-16 ***
lsales 0.16213 0.03967 4.087 6.67e-05 ***
lmktval 0.10671 0.05012 2.129 0.0347 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.5103 on 174 degrees of freedom
Multiple R-squared: 0.2991,

Adjusted R-squared: 0.2911

F-statistic: 37.13 on 2 and 174 DF, p-value: 3.727e-14

#**********b**********#
# Rename things to make them more accessible
> profits <- ceosal2$profits

# Fit the model
> fit2 <- lm(lsalary ~ lsales + lmktval + profits, data=ceosal2)

# Obtain values for equation
> summary(fit2)

Call:
lm(formula = lsalary ~ lsales + lmktval + profits, data = ceosal2)

Residuals:
Min 1Q Median 3Q Max
-2.27002 -0.31026 -0.01027 0.31043 1.91489

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 4.687e+00 3.797e-01 12.343 < 2e-16 ***
lsales 1.614e-01 3.991e-02 4.043 7.92e-05 ***
lmktval 9.753e-02 6.369e-02 1.531 0.128
profits 3.566e-05 1.520e-04 0.235 0.815
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.5117 on 173 degrees of freedom
Multiple R-squared: 0.2993,

Adjusted R-squared: 0.2872

F-statistic: 24.64 on 3 and 173 DF, p-value: 2.53e-13

#**********c**********#
# Rename things to make them more accessible
> ceoten <- ceosal2$ceoten

# Fit the model
> fit3 <- lm(lsalary ~ lsales + lmktval + profits +ceoten, data=ceosal2)

# Obtain values for equation
> summary(fit3)

Call:
lm(formula = lsalary ~ lsales + lmktval + profits + ceoten, data = ceosal2)

Residuals:
Min 1Q Median 3Q Max
-2.48792 -0.29369 0.00827 0.29951 1.85524

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 4.558e+00 3.803e-01 11.986 < 2e-16 ***
lsales 1.622e-01 3.948e-02 4.109 6.14e-05 ***
lmktval 1.018e-01 6.303e-02 1.614 0.1083
profits 2.905e-05 1.503e-04 0.193 0.8470
ceoten 1.168e-02 5.342e-03 2.187 0.0301 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.5062 on 172 degrees of freedom
Multiple R-squared: 0.3183,

Adjusted R-squared: 0.3024

F-statistic: 20.08 on 4 and 172 DF, p-value: 1.387e-13

#**********d**********#

# Find the sample correlation coefficient between the variables log(mktval) and
profits
> cor(lmktval, profits)
[1] 0.7768976