econ2P91_Lab4_SOLUTIONS_Winter2011

# econ2P91_Lab4_SOLUTIONS_Winter2011 - ECON 2P91 Business...

This preview shows pages 1–2. Sign up to view the full content.

ECON 2P91: Business Econometrics with Applications Winter 2011 Lab4 SOLUTIONS The objective of this week’s labs is to demonstrate the estimation of panel regressions using both “pooled OLS” and “Least Squares Dummy Variables.” What is the price elasticity of demand for wine in Canada? To answer this question, we obtained panel data for 10 Canadian provinces each of which was observed over 22 years (1980-2001). These data are reported in the Excel data file Wineconsumption.xls . For this data set, CONS denotes the annual wine consumption (litres) per capita (15 years and over) and PRICE denotes the Consumer Price Index (CPI) for Wine (1980=100). D1, D2, D3, …. .,D10 are binary variables representing the individual (provincial in this case) fixed effects. For example, D1=1 if the province is Alberta and D1=0 for all other provinces; D2=1 if the province is British Columbia and D2=0 for all other provinces. Similar interpretations apply to D2 (Manitoba), D4 (New Brunswick), D5 (Newfoundland), D6 (Nova Scotia), D7 (Ontario), D8 (PEI), D9 (Quebec) and D10 (Saskatchewan). As noted in class, double log regressions are most appropriate for obtaining the price elasticities directly. (a) Generate the logarithm of CONS and call it LCONS; also generate the logarithm of PRICE and call it LPRICE. (b) Distinguish between a balanced panel and an unbalanced panel. Identify whether this dataset is a balanced panel or an unbalanced panel. A balanced panel has no missing observations. i.e. each of the 10 provinces has exactly the same number of observations in a balanced panel. An unbalanced panel has some missing observations i.e. some provinces have more observations than others. This data set is clearly a balanced panel since each province has exactly the same number of observations (22). Hence the

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.