Data sources: The present study is based on primary and secondary data. Fifteen egg farmers, 6 arathdars, 10 wholesalers, 5 wholesaler cum retailer, 16 retailers and 3 farias were selected for interview. Primary data were collected in September 2011 through face to face interview with the selected respondents in the study area. Farm level data were collected from Sreepur Upazila of Gazipur district and traders’ level data were collected from five markets of Gazipur district. Secondary data were accumulated from FAO publication, official records, books, journals and from the various statistical year books. The weekly average wholesale prices of eggs of various markets like Dhaka, Gazipur, Rajshahi, Chittagong, Sylhet, Khulna and Mymensingh during 2000 to 2011 were collected from Department of Agricultural Marketing (DAM). Latter it was converted into monthly figures.
Analytical Techniques: Depreciation cost, Marketing efficiency, farmer’s net prices, price spread, farmer’s gross share and farmer’s net share were calculated by using slandered formula.
Market integration: Market integration was measured by co-integration method. The bulk of econometric theories have been based on the assumption that the underlying data process is stationary. Stochastic process is said to be stationary if its mean and variance are constant over time and the value of covariance between two time periods depends only on the distance or gap or lag between the two time periods and not the actual time at which the covariance is computed (Gujarati, 2003, p.797). In practice, most economic time series are nonstationary. Aapplying regression models to non-stationary data may arise the problem of “spurious or nonsense” correlation (Gujarati, 2003, p. 792). To overcome such problems, the concept of co-integration was used because it offers a means of identifying and hence avoiding the spurious. The underlying principle of co-integration analysis is that, although trend of many economic series show upward or downwards over time in a nonstationary fashion, group of variables may drift together.
Unit Root and Co-integrationTest: The individual price series were tested for the order of integration to determine whether or not they are stationary which is known as the unit root test (Gujarati, 2003, p.799). A number of tests for stationary are available in the literature; these include the Dickey-Fuller (DF) test (DickeyandFuller, 1979), the Augmented Dickey-Fuller (ADF) test (Dickeyand Fuller, 1981) and the Philips- Perron (PP) test (Perron,1988).
Spatial Price Relationship: To test the market integration, the following co-integration regression was run for each pair of price series:
Yit = α0 + α1 Yjt + εt
Where, Yi and Yj are price series of a specific commodity in two markets i and j, and εt is the residual term assumed to be distributed identically and independently. The test of market integration is straightforward if Yi and Yj are stationary variables but if the price series proved as non-stationary then we have to done another test (Engle-Granger test) Testing whether the variables are co-integrated is merely another unit root test on the residual. However, since the Yi and Yj are individually non-stationary, there is the possibility that the regression is spurious. The DF and ADF tests in the present context are known as Engle-Granger (EG) test whose critical values was provided by Engle-Granger (Ramakumar, 1998).