WEBVTT

00:01.190 --> 00:07.600
Previously, we have understood the basic
idea of our trek to. And in this video

00:07.600 --> 00:15.530
we are going to look at a couple of other options that we
can use without trek to and that would enhance its features

00:16.050 --> 00:21.100
So let me install the
dataset into my Stata file

00:22.400 --> 00:26.860
We are using the same data set
that we used in the previous video

00:27.410 --> 00:33.220
We described the data set. We know what
variables are there. We run our regression

00:34.530 --> 00:39.960
and now we want to report
the results in the world file

00:40.410 --> 00:45.270
but instead of our square, what
we want is adjusted R square

00:45.630 --> 00:54.640
Remember that previously, whenever we reported the the results,
it would show the R square value rather than the adjusted

00:54.640 --> 00:55.630
R square value

00:57.080 --> 01:06.310
but we are interested in the arts, a district Oscar
value. So the option that we use is a adjust our square

01:06.310 --> 01:11.200
and it would replace our square with
the district are square that moronic

01:13.340 --> 01:17.230
So now you can see we do not have
the R square instead of R square

01:17.230 --> 01:18.950
we have the adjusted R square

01:22.030 --> 01:26.370
So imagine, let's say I do not want
the R square or gestured our square

01:26.370 --> 01:32.380
Then the option that we use is no
are square. So when I use this option

01:32.380 --> 01:35.660
it would remove the R
square or adjusted R square

01:37.810 --> 01:46.480
So you can see that the other data is there, but the R
square value registered R square value had been removed

01:47.840 --> 01:52.910
Now, similarly, if I do not want to
report the number of observations

01:52.910 --> 01:57.110
then I use the no observation option

01:58.150 --> 01:59.740
So let me run this command

02:00.620 --> 02:04.240
So now you can see, I didn't
use the no our square option

02:04.240 --> 02:11.740
but I used the no observation option. So
it had deleted the number of observations

02:13.280 --> 02:22.670
Now in case I want to report the results in a side way
rather than having the coefficients and the standard error

02:23.170 --> 02:28.940
in the next row, I want them in two separate
columns. So I use this side of the option

02:29.100 --> 02:30.950
Let me show you how it looks

02:32.480 --> 02:37.940
So now you can see the coefficients are in one
column and the sternum errors are in their column

02:38.050 --> 02:43.720
But the issue is that we have parenthesis in
here and we do not want these parenthesis

02:45.350 --> 02:49.930
So we use the no parent. This
is option which is no parent

02:49.930 --> 02:56.850
And by using this option, still decide
way including the sideway option

02:56.850 --> 02:59.990
What we use we do is we
remove the parenthesis

03:01.270 --> 03:04.050
So now you can see that
there are no parenthesis

03:05.790 --> 03:09.880
Now by default, we just get the R
square and the number of observations

03:09.880 --> 03:15.510
the coefficients and the standard. But what
if we wanted to add some other statistics

03:15.510 --> 03:24.700
say for example, the F statistics or the degree of
freedom. So the option that we have is our stats

03:25.380 --> 03:29.060
And then within that option,
we we ride the legend

03:30.230 --> 03:36.310
So in this case, this is the F stats comma
the value that would be reported there

03:36.690 --> 03:40.380
And let me show you first and
then I would explain it further

03:42.760 --> 03:50.710
So now you can see that the F statistics, the P value of
the F statistics and the degree of freedom had been reported

03:51.400 --> 03:53.750
Let me go back to the command view

03:54.730 --> 04:01.130
Now how do we write this? I have mentioned
that we write the ad starts the option name

04:01.130 --> 04:08.590
open parenthesis. Then the legend which will be
reported in which will be written in the first column

04:08.960 --> 04:15.560
Then we need to tell the, the
option, the starter that how

04:15.560 --> 04:21.960
where these, in which variable it
is saved. So what Stata does it

04:21.960 --> 04:31.420
it saves the results of the previous regression in
some variables. And there are two ways of knowing what

04:31.420 --> 04:38.160
variable the Stata have used to save them. The
first one is to use the written list command

04:38.490 --> 04:45.240
And the second one is to go towards the
documentation of the regression command

04:45.740 --> 04:50.560
So let me show you both these
options. So first one is to use

04:51.830 --> 05:00.690
he returned list and it would
display the, the, the, the variables

05:01.480 --> 05:09.930
that had been used to store the results of previous
regression command or whichever command that we have used

05:09.930 --> 05:14.620
So we can see that the F
statistics is stored in E F

05:15.670 --> 05:24.720
So the number can be matched here.
The P value is stored in E and P

05:24.720 --> 05:30.350
in parenthesis and the degree of
freedom is stored in. Let me show you

05:31.440 --> 05:39.970
in this E D R. So we just need to copy them
and paste them in our air starts option

05:41.370 --> 05:47.140
The second is to use the help file
of the command that we are using

05:47.140 --> 05:52.510
So in this case, we are using the regress
command. If you are using the X direct

05:52.510 --> 05:57.840
then you type help X direct
press enter and scroll down

05:59.820 --> 06:08.870
to, to the area where it shows the store results and how
the this command stores the results that tells us the number

06:08.870 --> 06:16.960
of observations are stored in E N the model,
some of square is used in store in this variable

06:16.960 --> 06:21.710
We just need to use the name of
this variable in our command

06:23.260 --> 06:28.930
So there are other couple of shorthand
commands like no S E D T stat command

06:28.930 --> 06:35.110
And what no S C would do is it
would not report any external errors

06:35.750 --> 06:41.940
So what do we do is we, we
use the same command and

06:42.790 --> 06:47.390
we type no S A press
enter to open the file

06:48.380 --> 06:52.040
And now you can see the conditions
are there, the other details are there

06:52.040 --> 06:59.730
But the standard error not reported similarly, instead of standard
error, we can report the T stats by using the T stat option

07:01.430 --> 07:07.890
the P value by using the P value option, the
confidence interval or the standardized beta

07:09.070 --> 07:13.540
So these are shorthand options that
we can also use without right to

07:14.360 --> 07:20.160
moving forward. There is another option
which is called stat and what it does it

07:20.160 --> 07:27.860
it reports the statistics that we we
tell this outlet to command to report

07:28.660 --> 07:37.770
So in this case, I want the coefficient, the standardized beta, the
sternal error and the confidence interval to be reported in columns

07:40.400 --> 07:48.060
you can see that the variables data is reported
in columns. So we have the confidence interval

07:48.060 --> 07:53.710
the standard error, the standardized beat us
and the coefficient regression coefficients

07:54.300 --> 07:55.950
We can also tell stata

07:57.310 --> 08:06.020
That what level of confidence that we need to use. And the
command that we use for this is level and within parenthesis

08:06.020 --> 08:11.500
we write the conference in terrible. By
default, the confidence interval is 95

08:12.640 --> 08:15.380
but we can change it using this level command

08:16.530 --> 08:24.900
We can also change the significance level.
So by default, the significance level is 1%

08:25.310 --> 08:26.360
and 10%

08:27.140 --> 08:36.160
So instead of that, what we want is we want Starks to
be reported with just the 5% level of significance and

08:36.160 --> 08:38.050
the 1% level of significance

08:38.850 --> 08:47.740
By default, it would report the statistics
for the 10% 5% and the 1% significance level

08:47.970 --> 08:53.970
The option that we use is alpha and within
alpha, we retire the significance level

08:53.970 --> 09:01.680
So in this case, I want the 1% significance level
and the 5% significance level at Miranda's command

09:02.890 --> 09:07.310
So I have reported the T stats so
that you can make the comparison

09:07.310 --> 09:14.980
In this case, this this variable is significant
at 10%. But there's no static static reported

09:15.740 --> 09:21.090
in the outback to command because I have
asked the outbreak to commander to do so

09:21.790 --> 09:27.520
But we can see that this is significant at 5% and
we have one static and this is significant act

09:28.370 --> 09:37.580
one person and we have two static. And similarly, the legends had
the historic significance level in the notes had also been changed

09:38.400 --> 09:44.630
But instead of this, what I want is to be consistent
with my previous tables or with other researchers

09:44.630 --> 09:47.160
what I want is I want a three star

09:48.450 --> 09:53.440
For 1% level of significance and to
static for 5% level of significant

09:53.440 --> 09:56.500
And we have an option for
this in out back to command

09:57.310 --> 10:02.100
The option is that we use
the symbols. So in this case

10:02.100 --> 10:10.250
I want the symbol three statics for this ALF
option. And then I want to start for this ALF option

10:10.250 --> 10:19.360
Let me run this command. And now you can see previously
we had two stacks with this constant value and we had

10:19.360 --> 10:27.690
one start with this this value. And now it had been changed to
whatever the value of this direct number of tax or whatever symbol

10:27.690 --> 10:32.750
I want chatter to be and the
note had also been adjusted

10:32.750 --> 10:41.880
Similarly, sometimes they would want this direct to be reported
with the T stat values rather than with the coefficient values

10:42.530 --> 10:51.790
or with some other values. So what we use an option is
static and then we write the name of the statistic that

10:51.790 --> 10:56.120
we want it to be used with.
So let me run this command

10:56.380 --> 11:04.720
Now you can see that instead of the static with the confidence
we have the status with the T stat value or you can simply

11:04.720 --> 11:12.220
remove an hysterics. And for that, we use
the no Aztec option and let me see this

11:12.610 --> 11:20.000
So no statics had been reported. And lastly,
we are going to have this keep and drop option

11:20.190 --> 11:21.730
And what it does is

11:22.830 --> 11:31.760
it only reports the variable that we want it to. So the keep would
only keep the variables that are written within the parenthesis

11:32.210 --> 11:36.000
So although there are multiple variables in the regression

11:37.000 --> 11:43.980
there are these M P G, these four variables, but we
want to report the results of just MPG and truck

11:45.430 --> 11:54.610
So let me run this command. And now you can see although
the regression had been run on all those four variables

11:54.610 --> 12:00.450
but just the two variables are reported.
And this can come in handy when we have here

12:00.450 --> 12:08.950
dummy or country dummy or prosection dummies. And we just want
to report the other variables rather than those dummy values

12:09.430 --> 12:14.080
the same objective can be achieved
with this drop option. So we

12:14.080 --> 12:17.540
we write the name of the
variable that we want to drop

12:18.390 --> 12:23.260
So drop in previous case, we always
reported the MPG in the trunk

12:23.260 --> 12:28.120
So that means we need to drop the head
room and the displacement variable

12:28.120 --> 12:33.450
So they would be there in the regression command
but they would not be reported in the table