Thumbnail Image

Theses

Browse

20231
Reset filters
Subject: Agricultural Statistics and Informatics × Author: GAYAN, ANKUR JYOTI × Type: Thesis ×
Reset filters

Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    ThesisItemOpen Access
    INTERVAL ESTIMATION FOR DECISION MAKING IN CROP YIELD OF ASSAM
    (2023) GAYAN, ANKUR JYOTI; Saikia, Hemanta
    Crop yield is an important factor in agricultural sector because it is directlyrelated to food security, essential for economic development, important for employmentgeneration and has significant impact on export revenue. So estimation of crop yield isequally important because it helps in better planning for farmers, optimizing resourceallocation and forecasting market supply and demand. In this study crop yield of the stateof Assam was taken into consideration. Under this study, all the necessary data has beencollected from secondary sources, which include district level data from ICRISAT website,Statistical handbook of Assam and Economic survey of Assam. The objectives of the study are: 1. Estimation of parameters and their applications with inequalities for the normalpopulation 2. Application of the bootstrap method for the non-normal population to test thevariances. Simple linear regression was used to estimate prediction interval and it wasfound that for 23 districts of Assam yield of rice, pigeon pea, sesamum, rapeseed &mustard the prediction interval is true for 85.21%, 83.48%, 86.95% and 85.21%respectively for the period of 2018-2022. P-P plot was used to distinguish betweennormally distributed data and non-normally distributed data where it was found that foryield of rice, pigeon pea, sesamum, rapeseed & mustard there were 8, 5, 10 and 13 districtsof Assam respectively which were normally distributed. For normally distributed data theaverage rice yield, pigeon pea yield, sesamum yield, rapeseed & mustard yield ofHailakandi, Dibrugarh, Karbi Anglong and Jorhat respectively were the highest. Also,Chebyshev’s inequality was used to estimate confidence interval for normally distributeddata. At last bootstrapping method was used to test the equality of variances for non-normal data.