Design Choices: Should I use subsequent years or take a long difference?
Design Choices: Should I use subsequent years or take a long difference?
Loading saved threads...
Khanal · External communityPost link
External question — Economics Stack Exchange
Author: Khanal
Original post: https://economics.stackexchange.com/questions/61190
License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/
Adaptation: HTML converted to plain text; contact email addresses removed.
I need your help deciding on a design choice for my research. I want to look at the effect of change in exchange rate (
$D$
) exposure on the region's cumulative loan (
$Y$
) for farm operations of region
$i$
. I can either use 5-year intervals and only focus on census years and take the first difference as
$Y_{i,t} - Y_{i,t-5}$
and
$D_{i,t} - D_{i,t-5}$
, and conduct the analysis. I can also just take consecutive years and analyze with
$Y_{i,t}-Y_{i,t-1}$
and
$D_{i,t} - D_{i,t-1}$
. Of course, the number of observations will be larger for the second observation. However, apart from this specific advantage, what are the other things should I keep in mind when deciding on the main specification?
Thank you in advance.
Quote
Report
Nico Damascus · External communityPost link
External answer — Economics Stack Exchange
Author: Nico Damascus
Original post: https://economics.stackexchange.com/a/61198
License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/
Adaptation: HTML converted to plain text; contact email addresses removed.
You should use shorter intervals if you want to focus on shorter-term effects, and you should use longer intervals if you want to see longer-term effects.
Quote
Report
Post Reply
Quoted from Forex.com.bd-Editorial External question — Economics Stack Exchange Author: Khanal Source score (net votes, not local likes): 0 Original post: https://economics.stackexchange.com/questions/61190 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. I need your help deciding on a design choice for my research. I want to look at the effect of change in exchange rate ( $D$ ) exposure on the region's cumulative loan ( $Y$ ) for farm operations of region $i$ . I can either use 5-year intervals and only focus on census years and take the first difference as $Y_{i,t} - Y_{i,t-5}$ and $D_{i,t} - D_{i,t-5}$ , and conduct the analysis. I can also just take consecutive years and analyze with $Y_{i,t}-Y_{i,t-1}$ and $D_{i,t} - D_{i,t-1}$ . Of course, the number of observations will be larger for the second observation. However, apart from this specific advantage, what are the other things should I keep in mind when deciding on the main specification? Thank you in advance.
Checking account access…