University of zimbabwe faculty of social studies department of economics



Download 0.76 Mb.
View original pdf
Page16/45
Date16.12.2020
Size0.76 Mb.
#54545
1   ...   12   13   14   15   16   17   18   19   ...   45
taku dissertation
taku dissertation, taku dissertation
3.1.2 The Empirical Model
This study utilised empirical specifications by Muchapondwa (2008), Leaver (2004) and Townsend and Thirtle (1997). The study closely followed the ARDL model of the adapted
Nerlove (1958) log-linear supply function by Leaver (2004). Minor modifications were made by using a different time period and dropping some variables due to unavailability of data. The adoption of the logarithmic function allows coefficients to be directly interpreted as elasticities and also reduces problem of non-normality in residuals. The tobacco supply response econometric model is presented as follows


12 ln(TOUT
t
)=β
0

1
ln(TOUT
t-1
)+β
2
ln(PT
t-1
)+β
3
ln(MP
t-1
)+β
4
ln(ATG
t
)+β
5
SQ
t-
1

6
T+β
7
T
2

t
…………………………………………………………………………….....(3.2) where β
0
= regression intercept β
1
to β
7
= regression coefficients ln(TOUTt)= natural log of total tobacco output in tonnes ln(TOUT
t-1
)= log of lagged total tobacco output ln(PT
t-1
)= natural log of lagged tobacco price ln(MP
t-1
)= natural log of lagged maize price, expressed in US dollar terms per tonne ln(ATG
t
) = natural log of annual population of active tobacco growers SQ
t-1
= dummy variable for the years with sales quota taking the value of 1 in years in which sales quota applies and 0, otherwise T simple time trend (t for 1980 tot for 2015); T quadratic time trend (t for 1980 tot fort random error term with zero mean and constant variance.

Download 0.76 Mb.

Share with your friends:
1   ...   12   13   14   15   16   17   18   19   ...   45




The database is protected by copyright ©ininet.org 2024
send message

    Main page