Help required for variable forecast procedure

[Yahya Ghazali]

Dear All

I need some help with commands relating to earnings forecasts model:

EARN(t+1)=b0+ b1EARN(t)+b2BV(t)+b3CE(t)+e

The literature outlines the following steps to calculate the one-year-ahead forecast earnings (these steps are explained through my data example):

Step 1: For instance, the year ‘t’ is 2010 and we want to forecast the earnings (EARN (t+1)) for 2011 for each firm. So firstly, according to the literature, we run the dependent var (EARN) year t-1, i.e 2009 on the independent variables t-2, i.e 2008 and store the coefficients. Could you please help me with the codes for these?

Step 2: Next, multiply the obtained coefficients of independent variables from step 1 with the independent variables of the year 2010 to estimate the earnings (EARN t+1) for 2011.

Step 3: Repeat step 1 and 2 for other years i.e from 2011 until 2015 and then take the average estimated coefficient from the pooled cross-sections regression estimated each financial year from 2011 until 2015. Please provide the commands for this.

I will greatly appreciate if you could please help me with the codes for these.

Many thanks.

Yahya Ghazali


. dataex StockCode YEAR EARN BV CE if StockCode==11 & YEAR>=2009 & YEAR<=2014

copy starting from the next line ------- - ---------- --

* Example generated by -dataex-. To install: ssc install dataex
clear
input long StockCode float YEAR double(EARN BV CE)
11 2009  96933951.02  662304640.18   2771658.5
11 2010 174998534.79  875047708.94 19488536.34
11 2011 257461077.54 1131105960.98 12411814.98
11 2012 375422129.64 1503715002.59  4327597.65
11 2013 300840563.81 1803643379.74 12696883.02
11 2014 417498679.91 2075104749.13 13682765.56
end
format %ty YEAR

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. Enough data to work with would be helpful.

I'm confused by your questions. You've been on this list serve for 4 years but seem to be asking questions as if you have no understanding of Stata or statistics.

There are many ways to predict time series values. Read the time series documentation.