The study areas were selected at Rangpur, Thakurghaon, Comilla under the agro-ecological zones of Tista Meander Floodplain, Old Himalayan Piedmont Plain and Middle Meghna River Floodplain respectively. Data were collected from primary and secondary sources. Primary data were collected from face to face interview method using a pre-tested interview schedule by the researchers with the help of trained enumerators for the period of January - February, 2012. Secondary data were from government and non government organizations, such as Soil Resources Development Institute (SRDI), Bangladesh Bureau of Statistics (BBS), Food and Agriculture Organization (FAO), Department of Agricultural Extension (DAE) etc. Primarily 150 farmers from 9 districts (Panchaghar, Dinajpur, Thakurgaon, Nilphamari, Gaibandha, Rangpur, Lalmonirhut, Chandpur, Comilla) of Bangladesh were randomly selected and they were trained with modern technologies. At last data were collected from 50 selected farmers and compiled finally according to the greater regions. Rangpur region comprises 20 farmers from Nilphamari, Gaibandha, Lalmonirhut and Rangpur districts, Thakurgaon region comprises 15 farmers from Panchaghar, Dinajpur and Thakurgaon districts and Comilla region comprises 15 farmers from Chandpur and Comilla districts. For this study the Following input pattern was maintained timely: Different varieties of potato were cultivated like Diamont, Cardinal and Granola. Potato varieties were sown on 15 November, 2011 and harvested on 15 February, 2012 in each region. The spacing of potato was 60 cm x 25 cm for whole tuber and 50 cm x15 cm for cut tuber and Seed rate was1.5 t ha-1. (Satter et al., 2005). Chemical fertilizers like Urea, TSP, MoP, Gypsum, Zingsul (zinc sulphate), Petroboron (Boric acid), Petromag (Magnesium Sulphate) and Joiboshakti (Bone Meal) were applied at the rate of 350 kg, 220 kg, 270 kg, 120 kg, 7 kg, 6 kg, 50kg and 50kg per hectare (BARI, 2005). Irrigation was applied in two times; after 35 DAP and 65 DAP.
Analytical techniques: Both fixed cost and variable cost were taken to calculate the total cost of potato cultivation. Land use cost was calculated on the basis of per year existing lease value of land. Irrespective of potato varieties, the profitability of potato production was examined on the basis of gross margin, net return and benefit cost analysis. The collected data were edited, summarized, tabulated and analyzed to fulfill the objectives of the study. Tabular method using descriptive statistics was mostly used in this study. The following Cobb-Douglas production function model was used to estimate the contribution of factors of potato production.
Y = AX1b1X2b2 ------------------------- Xnbneui
The production function was converted to logarithmic form so that it could be solved by least square method
i.e. lnY= a + b1lnX1 + ……………………… + bnlnXn + Ui
The empirical production function was the following:
lnY = a + b1lnX1 + b2 lnX2 + b3lnX3 + b4lnX4 + b5lnX5 + b6lnX6 + b7lnX7 + Ui.
Where, Y = Yield (kg/ha); X1 = Human Labor (Man-day/ha); X2 = Land preparation cost (Tk/ha); X3 = Seed (kg/ha); X4 = Manure (kg/ha); X5 = NPK fertilizer; X6 = Insecticide cost (Tk/ha); X7 = Irrigation cost (Tk/ha); a =Intercept; b1, b2 ---- b7 = Coefficients of the respective variables to be estimated; and Ui = Error term.