Thumbnail Image

Theses

Browse

2016 - 20187
Reset filters
Subject: Statistics × End date: 2018 × Start date: 2010 × Thesis Type: M.Sc ×
Reset filters

Search Results

Now showing 1 - 7 of 7
  • Thumbnail Image
    ThesisItemOpen Access
    Random walk and ARIMAX modeling for cotton yield in western zone of Haryana
    (CCSHAU, 2018) Alisha; Verma, Urmil
    Crop yield models are abstract presentation of interaction of the crop with its environment and can range from simple correlation of yield with a finite number of variables to the complex statistical models with predictive end. The pre-harvest forecasts are useful to farmers to decide in advance their future prospects and course of action. The study has been categorized into three parts i.e. the fitting of Random Walk, ARIMA and ARIMAX models for cotton yield forecasting in Hisar, Fatehabad, Sirsa and Bhiwani districts of Haryana. The Random Walk and ARIMA models have been fitted using the time-series cotton yield data for the period 1980-81 to 2010-11 of Hisar and Sirsa districts and 1997-98 to 2010-11 of Fatehabad district. The fortnightly weather data have been utilized as input series from 1980-81 to 2016-17 for fitting/testing the Random walk/ARIMA with weather input i.e. ARIMAX models. Models have been validated using the data on subsequent years i.e. 2011-12 to 2016- 17, not included in the development of the models.The multiple linear regression models with crop condition term as dummy regressor were fitted for Bhiwani district as the cotton yield data being stationary in nature and showing non-significant autocorrelations was not suitable for ARIMA modeling. Though, the MA models were tried but the yield forecasts were beyond acceptable limits. Random Walk i.e. I(1) and ARIMA(0,1,1) for Hisar, Fatehabad and Sirsa districts have been fitted for pre-harvest cotton yield forecasting. Alternatively, the Random Walk models with exogenous input were tried by utilizing the fortnightly weather variables (viz., TMIN1, RF11, SSH3 and SSH4 over the crop growth period). Lastly, the ARIMA models with alternative combinations of weather variables were tried for fitting the ARIMAX models. Following the steps required in SPSS; ARIMA(2,1,0) for Hisar and Fatehabad and ARIMA(0,1,1) for Sirsa districts along with fortnightly weather variables (viz., TMAX5, RF7, SSH4 and RH4 over the crop growth period) as input were finalized as ARIMAX models for district-level cotton yield forecasting. The predictive performance(s) of the contending models i.e. Random Walk, ARIMA and ARIMAX models were observed in terms of the percent deviations of cotton yield forecasts in relation to the observed yield(s) and root mean square error(s) as well. The level of accuracy achieved by ARIMA model(s) with weather input was considered adequate for estimating the cotton yield(s) i.e. the ARIMAX models consistently showed the superiority over Random Walk and ARIMA models in capturing the percent relative deviations pertaining to cotton yield forecasts. The ARIMAX models performed well with lower error metrics as compared to the Random Walk and ARIMA models in all time regimes.
  • Thumbnail Image
    ThesisItemOpen Access
    Comparative performance of different ratio estimators of population mean
    (CCSHAU, 2018) Tanu; Manoj Kumar
    In this study, an attempt has been made to compare the performance of different ratio estimators. For the said purpose, ratio estimators by different researchers have been taken. Comparison of proposed estimators have been done pair wise over bias and mean squared error. Theoretical conditions were also developed when one estimator is better than the other. A total of forty six conditions were found on each bias and mean square error. Theoretical conditions were also compared using the empirical data set in which all the parameters required for the estimators were calculated. R-software code also developed to compare the bias, mean square error and percentage relative bias for different estimators. It was observed that estimator proposed by Subramani and Kumarapandiyan ( ere found be the best in term of bias, mean square and percentage relative bias with all the proposed estimators whereas estimator proposed by Kadilar and Cingi (2004) found to be the worst estimator empirically.
  • Thumbnail Image
    ThesisItemOpen Access
    Evaluation of G × E interaction in pearl millet using AMMI and GGE biplot analysis
    (CCSHAU, 2018) Mamata; Hooda, B.K.
    In the present study, the G × E interaction in pearl millet (Pennisetum glaucum L.) genotypes from three zones of India have been evaluated using the techniques of AMMI and GGE biplot analysis. AMMI ANOVA indicated that environment contributes maximum part of variation followed by genotype × environment interaction and genotype. Interaction principal component axes i.e. IPCA1, IPCA2 and IPCA3 were found to be significant in all three zones. On the basis of ASV, genotypes MH 2091, MH 2120 and MH 2114 while from stability index, genotypes MH 2085, MH 2105 & MH 2108 were found to be most stable for Zone A1, Zone A and Zone B respectively. From YSI and WI, genotypes MH 2101 & MH 2091 and genotypes MH 2109 & MH 2107 were found most stable with high yield for Zone A1 and Zone A respectively while for Zone B, genotypes MH 2123 & MH 2113 and MH 2123 & MH 2114 were found most stable with high yield on the basis of YSI and WI respectively. SI (%) indicated two groups of stable genotypes i.e. very low & low for Zone A1 and low & moderate for Zone A and Zone B, respectively. Spearman‟s rank correlation coefficient between YSI and WI for all the three zones showed that the two indices have almost equal performance in determining high yielding stable genotypes. From scatter plot of normalized grain yield and normalized ASV most stable (Zone A1(MH 2085),Zone A(MH 2120)- and Zone B(MH 2120)), high yielding (Zone A1(MH 2098),Zone A(MH 2129)- and Zone B(MH 2106)) and most stable with high yielding (Zone A1(MH 2101),Zone A(MH 2109)- and Zone B(MH 2123)) genotypes were obtained. Three major aspects of GGE biplots indicated mega environment analysis, test environment evaluation and genotype evaluation. GGE biplots found the most stable genotypes MH 2087, MH 2098 & MH2081 for Zone A1, MH 2106 & MH 2129 for Zone A and MH 2103, MH 2106 & MH 2119 for Zone B. MH 2101, MH 2107 & MH 2113 were found to be high yielding genotypes for Zone A1, Zone A and Zone B, respectively. MH 2091, MH 2111 & MH 2106 were found to be most stable with high yielding genotypes for Zone A1, Zone A and Zone B.
  • Thumbnail Image
    ThesisItemOpen Access
    Weather based rice yield prediction models for Karnal District of Haryana
    (CCSHAU, 2017) Deepankar; Aneja, D.R.
    The study carried out for Karnal district of Haryana was based on historical data for a period of 35 years (1980-81 to 2015-16). Rice yield data were collected from Statistical Abstract of Haryana and daily weather data were obtained from Central Soil Salinity Research Insitute (CSSRI), Karnal. Stepwise multiple regression technique has been applied for period 1981-82 to 2012-13 with yield as dependent variable and weather indices (artificial variables generated from weekly & fortnightly weather values) as independent variables. Another three years data (2013-14 to 2015-16) have been used for the validation of the models. The models based on maximum temperature (22 & 24 weeks) and no. of rainy days (22 & 24 weeks) are comparable with each other on the basis of adjusted R2, therefore on the basis of root mean square error the model based on no. of rainy days (24 weeks) having lowest RMSE (219.14) is chosen among all models based on individual weather variables. The actual forecasts using model based on no. of rainy days (24 weeks) for 2013-14 to 2015-16 years were 3207.11 kg/ha, 3318.09 kg/ha and 3322.61 kg/ha, respectively. The model based on joint effect of maximum temperature and relative humidity morning (22 weeks) forecasts rice yield were very close to the actual yields (per cent relative deviation ranging from 1.15 % to 7.6 %. The actual forecasts using maximum temperature and relative humidity morning for 2013-14 to 2015-16 years were, 3244 kg/ha, 3168.03 kg/ha and 3215.34 kg/ha, respectively.
  • Thumbnail Image
    ThesisItemOpen Access
    Biplots and their application in multi-environment trials data analysis
    (CCSHAU, 2017) Bhushan Kumar; Hooda, B.K.
    The present study was intended towards the general principle of construction of biplots and their use in analysis of genotype versus environment interactions data. Various types of biplots have been described for wheat yield data from five locations of Haryana and six locations of North India. The software UB biplot was used for construction of different types of biplots in general while GGEbiplotGUI software was used for solving different purposes in analysis of G×E data which specially analyze GE data. Principal component analysis, Multidimensional scaling and Nonlinear biplots were constructed to highlight methodologies used in construction of biplots. In particular, genotype plus genotype × environment interaction biplots were constructed to solve many of problems faced by breeder and agricultural scientists which generally use PCA biplot. Though, scientists and breeder exploit GE data in a number of modes in their breeding programs. Six types of GGE biplots indicating similarities-dissimilarities among genotypes, relationships among environments, evaluation of test environments, mega-environment analysis, evaluation of genotypes and ranking of genotypes were constructed by analysis of GE data.
  • Thumbnail Image
    ThesisItemOpen Access
    Logistic regression models for preharvest wheat yield estimation in western zone of Haryana
    (CCSHAU, 2017) Sudesh Rani; Verma, Urmil
    Timely and effective pre-harvest forecast of crop yield is important for advance planning, formulation and implementation of policies related to the crop procurement, distribution, price structure and import-export decisions etc. These are also useful to farmers to decide in advance their future prospects and course of action. The statistical modelling approaches viz., multiple linear regression, principal component analysis and ordinal logistic regression were used to achieve pre-harvest wheat yield forecasting in Hisar, Bhiwani, Sirsa and Fatehabad districts comprising the western zone of Haryana. The zonal weather-yield forecast models have been developed using data from 1978-79 to 2009-10 and the data from 2010-11 to 2014-15 were used for validation of the fitted models. Fortnightly weather data starting from 1st November to 1 month before harvest i.e. 1st fortnight of March over the period 1978-79 to 2009-10 were utilized for the model building. Data for the last 1 month of the crop season were excluded as the idea behind the study was to forecast yield at least 4-5 weeks in advance of the crop harvest. Year/time variable was included to take care of the variation between districts within zone as the weather data were not available for all the districts, however, the zonal model utilized the same weather information in the adjoining districts under the zone so that a longer series could be obtained in a relatively shorter period. The predictive performance(s) of the contending models were observed on the basis of Adj-R2, percent deviations of wheat yield forecasts in relation to the observed yield(s) and root mean square error(s) as well. The overall results indicate the preference of using prediction equations based on principal component scores/ probabilities of response categories obtained in logit analysis over the regression models using weather parameters as predictors. Trend yield (Tr) has been observed an important parameter appearing in all the models, indicating that most of the variability in yield is explained by Tr, which is an indication of technological advancement, improvement in fertilizer/ insecticide/ pesticide/ weedicide used and increased use of high yielding varieties. The other question was to see the usefulness of zonal weather-yield models and that provided quite satisfactory results pertaining to district-level wheat yield forecast(s) in the state. The percent relative deviations falling within tolerable limits favor the use of developed zonal weather-yield models for district-level pre-harvest wheat yield estimation in the western zone of Haryana.
  • Thumbnail Image
    ThesisItemOpen Access
    Spatial and temporal distribution of monthly rainfall in Haryana
    (CCSHAU, 2016) Nain, Mohit; Hooda, B.K.
    The analysis of monthly rainfall pattern of a region over a number of years is very useful for crop planning and irrigations scheduling. The present study focused on the probability distribution of monthly rainfall in various IMD rain gauge stations and identification of homogeneous rainfall regions in Haryana. The spatial and temporal distribution of monthly rainfall for the period 1970-2011, covering 27 rain gauge stations of Haryana has been studied. Probabilities of drought, normal and abnormal events for monthly rainfall were worked out. The results revealed that drought months are more probable than normal months while normal months are more probable than abnormal months. In case of yearly rainfall, normal years are more probable than drought and abnormal years. For examining the monotonic trend direction and magnitude of change over time, the Mann-Kendall test and Sen’s slope estimator tests were used. Increasing as well as decreasing trends were observed at various rain gauge stations. A significant decrease in annual rainfall was noticed at Ballabgarh and Thanesar while significant increasing trend was noticed at Sirsa only. In monsoon rainfall a significant decreasing trend was observed at Thanesar and Narnaul while significant increasing trend at Sirsa. Clustering of rainfall stations for monsoon period was done using Ward’s method applied on the common principal components scores (CPCs). The results indicated that there are 4 clusters of rain gauge stations having similar monsoon rainfall spread over Haryana. Cluster analysis of mean monthly rainfall was also performed using Ward’s method. The two analyses gave the patterns in close agreement.