hey could you help me solve the questions I just post. Thanks

(using MS Word or something similar). Insert the relevant part of your Stata log file into your
1. The Stata dataset traffic_S2016.dta contains 8 years of monthly data (for a total of 96
observations) from California on the following variables:

the number of traffic accidents (totacc)
the percentage of accidents which are fatal (prcfat ? since it?s a percentage, the maximum
it could theoretically be is 100)
a dummy variable indicating whether the statewide seat belt law was in effect (beltlaw)
a dummy variable indicating whether the statewide 55 mph speed limit was in effect
(spdlaw55)
the year the data was collected (year, which ranges from 1982 to 1989)
11 dummy variables for the month (feb, mar, apr, may, jun, jul, aug, sep, oct, nov, and
dec)
a. Using the ?browse? command in Stata, which lets you just eyeball the data, what month
and year did the seat belt law go into effect? When was the last month that the 55 mph
highway speed limit was in place? (1 point)
b. Run a regression of the variable totacc on beltlaw and spdlaw55. Do the sign of the
slope coefficients on beltlaw and spdlaw55 make sense to you? (1 point)
c. One big reason why there was a big positive slope on beltlaw is because there is a large
time trend in the number of cars on the road (and thus in the number of accidents).
Rerun the regression in part (b) but add a time trend ? just put the year variable in the
regression. What do the slopes on beltlaw, spdlaw55, and year mean here? (2 points)
d. Finally, run two regressions with prcfat as the dependent variable; these two regressions
should be identical to the two regressions you ran for parts (b) and (c), except you use
prcfat instead of totacc as the dependent (Y) variable. Note that the slope on beltlaw
became smaller in absolute value (i.e., closer to zero) when you added the time trend,
but the slope on spdlaw55 became larger in absolute value. Why? Think about the
correlation of those two variables with the time trend (2 points)

2. Suppose that quarterly consumption in dollars (CONS) is a linear function of the amount of
disposable income in dollars (YDISP) with a two-order lag:

CONSt 0 1YDISPt 2YDISPt 1 3YDISPt 2 u t
a. A macroeconomist estimates ?1, ?2 and ?3 to be 0.3, 0.4 and 0.1, respectively. What is
the estimated effect of a one-time increase of 1000 in YDISP on CONS in this quarter,
and for the next 3 quarters following that (by ?one-time increase?, we mean that in the
next quarter, YDISP goes back down to its original level)? (2 points)
b. What is the effect of a permanent increase in YDISP of 1000 on the long-run average of
CONS? (2 points)