Forecasting using SAS Forecast Server, SAS/ETS, and more

Changes in returns relative to the sample-end date

Reply
Contributor
Posts: 57

Changes in returns relative to the sample-end date

[ Edited ]

Dear All,

 

I am simply trying to test whether the return of dependent variable increases/decreases after the sample end date (December 2001). For example, my dependent variable is L_S (long-short portfolio) of stock i at time t and the independent variable is post sample dummy for stock i at time t. Post sample dummy is equal to one if the date is after 2001, else zero.

L_S(it) = alphai + beta1 post sampel dummy(it) + e(it)

 

Can you please help me how to run the regression test for changes in returns relative to the sample end date (December 2001)? I have provided sample data too.

data have;
infile cards expandtabs truncover;
input stock	date : yymmn6. L_S post_sample	;
format Date yymmn6.;
cards;
14	199901	14.0238	0
15	199901	11.9454	0
16	199901	9.0066	0
14	199902	1.5165	0
15	199902	-1.9619	0
16	199902	0.3936	0
14	199903	1.9127	0
15	199903	-0.2142	0
16	199903	-6.9887	0
14	199904	1.9835	0
15	199904	1.1102	0
16	199904	1.1719	0
14	199905	1.771	0
15	199905	3.2528	0
16	199905	-0.7609	0
14	199906	0.2883	0
15	199906	1.174	0
16	199906	-0.3383	0
14	199907	3.1988	0
15	199907	2.2924	0
16	199907	-2.6332	0
14	199908	0.3939	0
15	199908	-0.1062	0
16	199908	-2.1077	0
14	199909	2.2333	0
15	199909	2.3855	0
16	199909	-0.9922	0
14	199910	0.0906	0
15	199910	-1.7595	0
16	199910	1.8265	0
14	199911	12.0945	0
15	199911	13.1003	0
16	199911	11.7012	0
14	199912	13.1348	0
15	199912	13.0311	0
16	199912	5.6075	0
14	200001	16.5104	0
15	200001	13.4878	0
16	200001	1.6201	0
14	200002	24.751	0
15	200002	22.8749	0
16	200002	9.1362	0
14	200003	-7.1384	0
15	200003	-5.9835	0
16	200003	-7.9094	0
14	200004	-13.1302	0
15	200004	-12.6789	0
16	200004	-12.3907	0
14	200005	-7.179	0
15	200005	-8.9204	0
16	200005	-12.4663	0
14	200006	9.8039	0
15	200006	10.4251	0
16	200006	4.0955	0
14	200007	-6.7036	0
15	200007	-6.3405	0
16	200007	-2.261	0
14	200008	1.7284	0
15	200008	2.8166	0
16	200008	-0.1966	0
14	200009	-6.1185	0
15	200009	-6.2332	0
16	200009	-8.9549	0
14	200010	-9.5955	0
15	200010	-8.6117	0
16	200010	-7.8711	0
14	200011	-15.9693	0
15	200011	-14.1019	0
16	200011	-17.3939	0
14	200012	-11.7599	0
15	200012	-11.0496	0
16	200012	-16.0614	0
14	200101	28.428	1
15	200101	24.3161	1
16	200101	24.8696	1
14	200102	-10.0572	1
15	200102	-9.8009	1
16	200102	-14.7445	1
14	200103	-8.1778	1
15	200103	-6.6534	1
16	200103	-12.8975	1
14	200104	3.5659	1
15	200104	4.775	1
16	200104	7.0148	1
14	200105	3.0865	1
15	200105	3.0205	1
16	200105	2.6813	1
14	200106	-3.0407	1
15	200106	-1.0529	1
16	200106	6.5331	1
14	200107	-7.8688	1
15	200107	-6.0558	1
16	200107	-6.5671	1
14	200108	-5.2509	1
15	200108	-4.9021	1
16	200108	-4.7708	1
14	200109	-6.6907	1
15	200109	-7.7812	1
16	200109	-6.3164	1
14	200110	8.0976	1
15	200110	10.2051	1
16	200110	6.8036	1
14	200111	4.6386	1
15	200111	4.0607	1
16	200111	9.7786	1
14	200112	2.4793	1
15	200112	3.2357	1
16	200112	6.4167	1
14	200201	0.3871	1
15	200201	0.1048	1
16	200201	-3.0061	1
14	200202	-7.5435	1
15	200202	-7.1932	1
16	200202	-10.9996	1
14	200203	1.1006	1
15	200203	3.6903	1
16	200203	-2.3129	1
14	200204	-6.706	1
15	200204	-4.6423	1
16	200204	-7.2183	1
14	200205	-3.319	1
15	200205	-3.5954	1
16	200205	-4.587	1
14	200206	-5.9677	1
15	200206	-6.8817	1
16	200206	-4.1095	1
14	200207	-4.3103	1
15	200207	-5.4936	1
16	200207	-1.6184	1
14	200208	-0.783	1
15	200208	-1.9367	1
16	200208	-3.7291	1
14	200209	-6.8038	1
15	200209	-4.4041	1
16	200209	-4.2089	1
14	200210	4.7228	1
15	200210	4.2958	1
16	200210	6.1159	1
14	200211	14.8031	1
15	200211	11.6538	1
16	200211	9.2207	1
14	200212	-7.44	1
15	200212	-5.6201	1
16	200212	-6.7057	1
14	200301	2.7663	1
15	200301	1.9519	1
16	200301	0.2942	1
14	200302	-0.5867	1
15	200302	-0.6244	1
16	200302	-3.116	1
14	200303	0.8218	1
15	200303	0.8057	1
16	200303	0.8502	1
14	200304	5.9883	1
15	200304	4.1283	1
16	200304	8.4368	1
14	200305	13.8637	1
15	200305	11.1063	1
16	200305	4.3414	1
14	200306	4.0383	1
15	200306	4.8102	1
16	200306	-0.4886	1
14	200307	4.8519	1
15	200307	4.2237	1
16	200307	3.1779	1
14	200308	2.7659	1
15	200308	3.2586	1
16	200308	-0.6638	1
14	200309	5.2582	1
15	200309	3.2569	1
16	200309	-0.4464	1
14	200310	2.961	1
15	200310	3.1561	1
16	200310	1.5347	1
14	200311	0.5451	1
15	200311	1.0336	1
16	200311	-4.545	1
14	200312	0.6783	1
15	200312	-0.7552	1
16	200312	-5.7656	1
run;

Thanks in advance for your help.

 

Best Regards,

 

Cheema

Respected Advisor
Posts: 4,648

Re: Changes in returns relative to the sample-end date

Assuming that e(it) is iid normal:

 

proc glm data=have;
class stock post_sample;
model l_s = stock|post_sample / ss3;
output out=res r=l_s_res;
run; quit;

proc univariate data=res normal;
var l_s_res;
probplot / normal(mu=0) square;
run;
PG
Contributor
Posts: 57

Re: Changes in returns relative to the sample-end date

Thanks for it, but it seems bit different than what I am looking for. I might not have explained it well. Please find the attached Table from one of the paper from Journal of Finance, I want to run the similar regression. Thanks for your help.

Super User
Posts: 9,681

Re: Changes in returns relative to the sample-end date


Sorry. I can not help you.
Maybe you should take a look at PROC PANEL

Contributor
Posts: 57

Re: Changes in returns relative to the sample-end date

Thanks, I wrote a code using process panel but it does not estimate change, it just estimate the relationship between dependent and independent variables. I will ask the authors of the paper, they might be able to help me. Thanks my friend.
Ask a Question
Discussion stats
  • 4 replies
  • 150 views
  • 2 likes
  • 3 in conversation