High P Values

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

darmanin
Posts: 1
Joined: Mon Jan 14, 2019 12:01 pm

High P Values

Postby darmanin » Mon Jan 14, 2019 12:24 pm

I am in the process of carrying out OLS for the following data.
Date GDP FDI Stock GFCF Inflation Lending Tertiary Trade Balance
2004 4,852.40 12,473.72 999.60 2.80 5.30 19.90 -1,246.10
2005 5,149.30 32,818.49 1,132.90 3.00 5.50 23.40 -1,435.10
2006 5,386.10 46,841.25 1,183.20 2.80 5.70 22.70 -1,321.20
2007 5,757.50 75,916.03 1,288.20 1.30 6.20 23.10 -1,461.80
2008 6,128.70 84,125.37 1,203.10 4.30 5.90 23.50 -2,109.50
2009 6,138.60 86,902.98 1,114.80 2.10 4.50 24.00 -1,928.90
2010 6,599.50 97,118.90 1,411.60 1.50 4.60 25.80 -2,021.00
2011 6,840.00 112,950.15 1,228.40 2.70 4.70 28.90 -2,117.00
2012 7,168.00 125,443.22 1,299.80 2.40 4.70 29.60 -2,249.70
2013 7,641.70 133,283.47 1,334.90 1.40 4.70 32.80 -2,276.50
2014 8,467.50 142,774.50 1,465.40 0.30 4.20 34.10 -3,491.70
2015 9,524.60 153,409.38 2,409.40 1.10 3.70 34.90 -2,811.50
2016 10,180.10 163,480.59 2,483.80 0.60 3.70 40.40 -2,788.40
2017 11,139.70 172,595.84 2,351.60 1.30 3.50 42.00 -2,769.30

GFCF is not normally distributed so the LOG was calculated.
ADF and PP were carried out on all variables GDP, FDI, LOGGFCF and Tertiary resulted as being non-stationary so these were differentiated in order to become stationary.

OLS Regression was then carried out on the following data



Below are the results of my OLS regression. As you can see Prob values are quite high and I believe something is wrong.

Dependent Variable: DGDP
Method: Least Squares
Date: 01/14/19 Time: 19:51
Sample (adjusted): 2005 2017
Included observations: 13 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C -7.696630 836.6242 -0.009200 0.9930
DFDI 0.017970 0.013005 1.381771 0.2163
DGFCF 614.6282 389.1661 1.579347 0.1653
INFLATION 36.16991 80.24407 0.450749 0.6680
LENDING -125.4424 140.3518 -0.893772 0.4059
DHUMANCAPITAL -17.92251 38.69744 -0.463145 0.6596
TRADEBAL -0.354756 0.152027 -2.333511 0.0584

R-squared 0.807228 Mean dependent var 483.6385
Adjusted R-squared 0.614456 S.D. dependent var 307.0603
S.E. of regression 190.6605 Akaike info criterion 13.64260
Sum squared resid 218108.5 Schwarz criterion 13.94680
Log likelihood -81.67689 Hannan-Quinn criter. 13.58007
F-statistic 4.187475 Durbin-Watson stat 1.713165
Prob(F-statistic) 0.052519

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: High P Values

Postby startz » Mon Jan 14, 2019 2:15 pm

You only have 13 observations. Probably nothing is wrong.


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 14 guests