Thumbnail Image

Theses

Browse

20183
Reset filters
Subject: Agricultural Statistics and Informatics × End date: 2018 × Start date: 2018 × Type: Thesis ×
Reset filters

Search Results

Now showing 1 - 3 of 3
  • Thumbnail Image
    ThesisItemOpen Access
    Multivariate analysis for the classification of locations using soil parameters in central districts of Kerala
    (Department of Agricultural Statistics, College of Agriculture, Vellayani, 2018) Muhsina, A; Brigit Joseph
    The research entitled ‘Multivariate analysis for the classification of locations using soil parameters in central districts of Kerala’ was conducted with the objectives to find out appropriate multivariate method to classify the locations based on soil fertility parameters and also to find out the interrelations among the soil parameters. Data on thirteen soil parameters of 17 panchayats (678 samples) in Ernakulam (EKM) and 28 panchayats (789 samples) in Kottayam (KTM) as per “Revolving Fund Mode Project- Soil Testing Lab” of Department of Soil Science and Agricultural Chemistry, College of Agriculture, Vellayani were subjected to various multivariate techniques. Initial data analysis using box plot technique was carried out to remove the outliers present in soil fertility parameters of both districts. Descriptive statistics of each soil parameters in the different panchayats in both districts were worked out and it was found that pH ranged from 4.2-5.8 in EKM with a coefficient of variation (CV) of 10.6 per cent, whereas that of KTM ranged from 4.61-5.95 with a CV of 11.28 percent. Oxidisable organic carbon (OC) (range: 1.23-2.47 %), phosphorus (range: 16.51-113.59 kg ha-1) and potassium (range: 45.86-570.21 kg ha-1) content had a CV of 46.84, 78.80 and 86.54 percent respectively in EKM while KTM showed a variation of 56.92, 79.24 and 60.21 percent respectively for OC (range: 0.98-3.39 %), P (range: 16.89-104 kg ha-1) and K (range: 179.14-580.64 kg ha-1). Multivariate analysis of variance (MANOVA) was done in order to test whether the panchayat means were significantly different with respect to 13 soil parameters. Results of MANOVA revealed that there was significant difference between the mean vector of 17 panchayats in EKM and 28 panchayats in KTM. Principal Component Analysis (PCA) generated five PCs which together accounted for 80 per cent variation in data of EKM and in KTM. Sampling adequacy for PCA and factor analysis was tested by Kaiser-Meyer-Olkin measure and it was found that sample sizes were adequate to conduct PCA in EKM (0.571) and KTM (0.464). Correlation among the variables was tested using Bartlett’s chi square test of sphericity and it was found that variables were correlated to each other. Variables were plotted using factor loadings and it was observed that EC, S and B had high positive loadings on factor 1 and 2 in EKM while none of the variables had positive loadings on F1 and F2 in KTM. Score plot obtained from PCA in EKM showed that Chengamanadu, Manjapra and Thirumarady panchayats had high content of available S and B. In Kottayam, Score plot drawn showed that Cu and Zn were predominant in Akalakkunnam, Kadaplamattom, Meenachil, Melukavu, Poonjar and Ramapuram panchayats. Hierarchical clustering (HC) and K –means clustering were performed to group the panchayats in both districts based on soil fertility status and thereafter comparison of various clustering procedures was done using Davies – Bouldin (DB) index. Different dissimilarity measures- Euclidean, squared Euclidean, Chebychev distance and Mahalanobis D2 were determined and single linkage, complete linkage and average linkage methods were adopted under these measures. The results in EKM showed that Mahalanobis D2 was the better clustering procedure with seven clusters (DB index: 0.120) followed by average linkage method under Euclidean distance (DB index: 0.306) with seven clusters. Manjapra (C6) and Keerampara (C7) remained as individual clusters. Keerampara had strongly acidic soils (pH -5.17) with high available Mg (158 mg kg-1) while Manjapra soils had low Mg availability (19 mg kg-1) and high S content (57 mg kg-1). Kakkad, Kalady and Vengoor came under C1 which possessed approximately same EC (0.15-0.19 dS m-1), OC (2-2.4%) and Mg (71-73 mg kg-1) content. Chengamanadu and Vengola came under C3 while Ayyampuzha and Mudakkuzha came under C4. Clustering of panchayats in KTM using Mahalanobis D2 resulted in better clusters as DB index was 0.120. Twenty eight panchayats in KTM were classified into 8 clusters. Only Akalakkunnam panchayat remained as a separate cluster with pH less than 4.8 and EC 0.08 dS m-1. Aymanam and Elikulam came under a cluster while Pallikkathode and Puthuppally came under another cluster. Bharananganam, Chempu and Melukavu came under a cluster as they had similarity in pH (5.35 -5.48), Mg (46-49 mg kg-1), S (23-27 mg kg-1) and Fe (23-29 mg kg-1). Inter relations among the variables were determined by using Pearson’s correlation and it was found that Ca, Fe and P was positively and significantly correlated with EC in KTM whereas S and P were positively related with EC in EKM. Mg and B were negatively related (-0.495). Similarly Fe and Mn had a negative correlation with each other (-0.467) in KTM district. Ca and B was negatively correlated in EKM (-0.525) which indicated the antagonistic effect. The results of the study indicated that Mahalanobis D2 gave the optimum clusters in both EKM and KTM followed by Euclidean distance using average linkage.
  • Thumbnail Image
    ThesisItemOpen Access
    Trends in production and bienniality of coconut (cocos nucifera L.) var.wct.
    (Department of Agricultural Statistics, College of Agriculture, Vellayani, 2018) Fallulla, V K; Vijayaraghava Kumar
    The study entitled "Trends in production and bienniality of coconut (Cocos nucifera L.) var.WCT" was carried out based on data on the number of nuts harvested from 525 WCT palms planted in 1966 at Coconut Research Station, Balaramapuram with its five to six harvests per year, for a period of 25 years viz. 1993 to 2017. The objectives of the study are to identify the extent of bienniality, variations in repeatability and type of yield fluctuations over years and over different harvests. The effect of meteorological factors like rainfall, number of rainy days, maximum and minimum temperature and wind velocity of the above period also formed part of this study. Initial data analysis using box plot technique was carried out to remove the outliers present in the data. The number of nuts produced by a palm in an year was found to be 68.5 with an overall standard deviation of 37.83 nuts. A plot of the average number of nuts produced in an year against the growing periods showed a steady decrease in yield from 2012 onwards (i.e. after 50 years of planting). Preliminary statistical analysis by applying Analysis of variance (ANOVA) for the number of nuts produced by individual palms revealed high significant difference between different palms with respect to each harvest and also with respect to each year. Pearson’s correlation coefficient between yield data of different harvests in an year as well as previous years were estimated and a significant correlation were observed for the previous harvest and rest of the coefficient were non significant. Statistical tools in respect of graphical, parametric and non parametric approaches were tried as an attempt to detect and quantify the biennial bearing tendency. Graphical approach confirmed biennial bearing tendency among different years as well as among different harvests. The parametric study was carried out using orthogonal contrasts developed by Saraswathi (1983). This method used four F ratios F1, F2, F3 and F4 , the significance of which provide biennial tendency and time-trend each for four year periods. F1 ratio is used to test the biennial tendency under the assumption of absence of time trend and then confirmed by F2 ratio. F3 is used to test time trend effect under the assumption of absence of biennial effect. This assumption confirmed by F4 ratio. For the period 1997-2000, F1 was found to be significant at 5 per cent level indicating biennial tendency for this period in the absence of time trend, which was then confirmed using F2 criterion. But this method didn’t confirm bienniality for other periods The non parametric approach using biennial bearing index ‘B’ (Hoblyn et al., 1936) was made for the period of 1993-2017. The ‘B’ factor was based on 23 pairs of successive signs positive or negative indicating fall or rise in yield over continuous years for each of the palms. A test of significance of bienniality was obtained by calculating the binomial probabilities. Number of successive change in signs of 16 or above for this period indicate significant departure from the equiprobable hypothesis. Therefore a palm showing a B factor equal to or higher than 16/23 can consider as significantly biennial in bearing; and on this basis 41.1 per cent of the palms were found to be biennial in bearing. Intensity or degree of crop fluctuations was measured by the ‘I’ factor (Hoblyn et al., 1936). All palms showed an intensity of crop fluctuations less than 50 per cent; of which in 81.8 per cent, the intensity ranged from 20 to 30 per cent. A zero percent ‘I’ indicates regular bearing or no alternate bearing behavior. Regular bearing was not observed for any of the palms. 100 per cent I indicates strict alternate bearing behavior. No palms were found to be strict in alternate bearing also. Maximum number of palms were found to exhibit the biennial bearing pattern but are not strict ( 100 per cent) in bienniality. Spearman’s rank correlation coefficients were calculated for all 23 pair of alternate years and all 24 pair of adjacent years. For palms possessing biennial tendency the coefficients for alternate years should be higher than that of adjacent years and this is tested by rank sum test (Z). The Z value was found to be non significant for the period 1993-2017 indicating no strict alternate bearing behavior in the selected palms. As production is found to be in a a steady decrease from 2002 onwards ‘Z’ is separately estimated for the period 1993-2001, and found to be significant for this period indicating alternate bearing behavior for this period. Repeatability was estimated for number of nuts per tree using ANOVA estimator for different periods. While considering the whole period 1993-2017 and 2013-2016 repeatability coefficient was very low 0.13 and 0.06 respectively with variances 0.00015 and 0.00052 respectively. High estimate of repeatability, 0.397, 0.355 respectively were observed for the period 1993-1996 and1997-2000. Correlation between climatic factors in the current year, previous year, two years before and three years before with the production of nuts for the current year were estimated and were not significant except for the minimum temperature of the current year. It indicated that the parameters of annual climatic factors were not adequate to explain the temporal variation in yield. However Correlation between number of days without rain in summer (dry spell ) and yield in succeeding season of next year was found to be -0.43 which is negatively significant, showing this factor will inversely affect the yield of the next harvest. Bienniality also found to be not directly influenced by the climatic factors. A linear regression model with high R2 value, 0.98 were fitted with current year yield as dependent variables and previous year yield, Number of trees in the ‘on’ phase, Rainy days and Wind velocity as independent variables.
  • Thumbnail Image
    ThesisItemOpen Access
    Multivariate clustering techniques- a comparison based on rose (rosa spp.)
    (Department of Agricultural Statistics, College of Agriculture, Vellayani, 2018) Arya, V Chandran; KAU; Vijayaraghava Kumar
    The study entitled “Multivariate clustering techniques – a comparison based on rose (Rosa spp.)” was undertaken to compare different clustering techniques, to identify the suitable technique for different types of qualitative and quantitative data and to illustrate the procedures using data based on a field experiment on rose (Rosa spp.). Data on quantitative and qualitative traits collected from a field experiment on “Characterization and genetic improvement in Rose (Rosa spp.) through mutagenesis” done during 2014-2017 at College of Agriculture, Vellayani and Regional Agriculture Research Station (RARS), Ambalavayal, Wayanad was used for the study. Twenty five cultivars each coming under the Hybrid Tea and Floribunda groups of rose were evaluated for the study. There were nine quantitative characters and three qualitative characters. Statistical studies were carried out with the help of statistical packages SPSS, STATA, SAS, R and NTSYS. Preliminary statistical analysis by applying Analysis of variance (ANOVA) for all quantitative characters under study revealed significant difference among different genotypes with respect to each character. Multivariate analysis of variance (MANOVA) was carried out to test the significance of varietal means for each group. The results indicated difference among the cultivar means for both groups with respect to all quantitative characters. Linear discriminant function developed using nine quantitative characters for each of the groups were used to elucidate the differences between them. The average score obtained was 11.01 for the Hybrid Tea type and – 2.34 for Floribunda type with an overall average of 4.38. Discriminant function analysis reassured the difference between the two groups under study. Cluster analysis on Hybrid Tea type and Floribunda type were performed for quantitative, qualitative and mixed data. Association measures used were Euclidean distance, Squared Euclidean, Chebychev distance, City Block distance and Mahalanobis D2 for quantitative data, Jaccard, Dice, Simple matching and Hamann’s coefficient for qualitative data and Gower’s measure for mixed data. Different methods such as single linkage, complete linkage, Unweighted Pair Group Average Method (UPGMA), Weighted Pair Group Average Method (WPGMA), Unweighted Pair Group Centroid Method (UPGMC), Ward’s method, modified Tocher method, k means clustering and Principal Component Analysis (PCA) were adopted for the clustering of cultivars. Optimum numbers of clusters were determined by Pseudo t2 statistics for hierarchical clustering and by Pesudo F statistics for k means clustering. SD ( Scatterness- Distance) index was used to test validity of clustering based on quantitative data. Clustering based on qualitative data was carried out using seven characters, three of which are qualitative traits and all others are quantitative characters converted to qualitative traits. Jaccard and Dice coefficient were used for binary data while Simple matching and Hamann’s were used for multi-state data. The result of different clustering techniques based on Squared Euclidean distance gave approximately the same result as that of Euclidean distance. The Jaccard and Dice coefficients were found to be very similar, so that there was no difference in topology of dendrogram but only in branch length. Clustering pattern under Simple matching and Hamann’s coefficient provided were of similar type. For both groups among all the clustering methods, single linkage clustering under different distance measures tends to create a set of one or two clusters including majority of the genotypes and the remaining genotypes are single or two member clusters. Single linkage clustering tends to produce long chain types clusters as opposed to bunched clusters. On the other hand, the single linkage algorithm suffers chaining effect. Among other clustering algorithms, complete linkage method and Ward’s clustering method showed similar results under Squared Euclidean distance. UPGMA, WPGMA and UPGMC methods under Squared Euclidean method gave comparable results. Clustering using UPGMA and WPGMA method gives almost same clustering pattern under different distance measures for qualitative and quantitative data. Results obtained from k means clustering are comparable with results obtained from hierarchical clustering except for single linkage clustering. A certain degree of similarity was observed between k means and D2 analysis but not to up that between other clustering methods. Under Hybrid Tea genotypes, H16 (Mary Jean) formed a single cluster under single linkage method using different distance measures for quantitative, qualitative and mixed data analysis. Under complete linkage method H7 (Alaine Souchen) and H25 (Josepha) came under same cluster, in clustering based on quantitative and qualitative characters. H22 (Mom’s Rose) and H23 (Lois Wilson) came under same cluster in clustering based on complete linkage, UPGMA and WPGMA except under Hamann’s coefficient. These came under the same cluster under D2 analysis also. Among Floribunda genotypes F2 (Tickled Pink) and F5 (Princess de Monaco) were included in the same cluster under UPGMA method for both quantitative and qualitative data. F1 (Versailles) and F24 (Golden Fairy) also came under the same cluster except for multistage distances under UPGMA. Clustering based on mixed data gave approximately the same results as that of quantitative data under different clustering algorithms except for single linkage clustering. Comparison using SD index indicated high index value for clustering based on Gower’s measure. Comparison among single linkage, complete linkage and Average linkage under different association measures using SD index were carried out. Average linkage method under Squared Euclidean was found to be the best for both type with SD index 0.651 for Hybrid Tea and 0.659 for Floribunda type. Clustering pattern observed from score plot of PCA is comparable with the pattern obtained from quantitative data especially with D2 analysis. Contribution of characters towards variance obtained D2 analysis and PCA showed similar results. From the study it is possible to compare different methods and exclude inappropriate methods. Groups formed from modified Tocher method and PCA are different from other methods. SD index indicated that UPGMA under Squared Euclidean distance is the best for quantitative data.