Showing papers by "Indian Agricultural Statistics Research Institute published in 2009"

Second-Order Response Surface Model with Neighbor Effects

Abstract: This article considers the second-order response surface model in which the experimental units, i.e., plots experience the neighbor effects from immediate left and right neighboring plots assuming the plots to be placed adjacent linearly with no gaps. Conditions have been derived for the estimation of coefficients of second-order response surface model. A method of constructing designs for fitting second-order response surface in the presence of neighbor effects has been developed. The designs so obtained are found to be rotatable.

Non-linear statistical models on estimation of maximum size of Tor putitora (Hamilton) in different aquatic environments

Abstract: Tor putitora is one of the most important coldwater fish species. The population of this fish species has declined sharply in the recent past and is threatened with multifaceted dangers. As the size of fish plays an important role in fish stock assessment, in the present investigation, an attempt has been made to estimate the maximum size of Tor putitora in different aquatic environments by using nonlinear statistical models. We can expect a maximum length of approximately 3097 mm and 2994 mm for Tor putitora in the aquatic environments of Kumaun lakes and Gobindsagar reservoir respectively. It is seen that the estimated maximum size of Tor putitora in both environments are well acceptable in view of the reported maximum sizes in India and abroad.

Approach for Mining Multiple Patterns from Clusters.

Abstract: Approach for multiple pattern extraction from obtained individual clusters is presented in this paper. Pattern extraction supports the end users in understanding the cluster concept. Pattern discovery approach uses reduct from rough set theory to find out non-significant attributes in pattern formation. These non-significant attributes (reduct) are removed and remaining attributes are ranked for their significance in the cluster. Multiple pattern formulation approach uses ranked attributes and results in generation of meaningful, concise cluster patterns. Applicability of the approach is demonstrated using soybean disease dataset from machine learning repository. Objective of applying proposed approach on soybean disease clusters is to obtain the different disease patterns.

An Empirical Investigation on Classical Clustering Methods

Abstract: Five classical clustering methods: four hierarchical -single linkage, average-between linkage, average-within linkage, Wards - and one non-hierarchical - k-means - using five different distance measures: squared Euclidean, city block, Chebychev’s, Pearson correlation and Minkowski have been compared on the basis of simulated multivariate data on paddy crop genotypes. The performance of different clustering methods was compared based on the average percentage probability of misclassification and its standard error. The performance of different hierarchical clustering methods varied with distance measures used and it was found that squared Euclidean performed best among the five distances followed by city block distance in majority of cases. Among the five methods, the Ward’s method performed best with least average percentage probability of misclassification followed by non-hierarchical k-means method irrespective of the sample size. Among the different distance measures used under hierarchical clustering methods, the squared Euclidean distance showed least average percentage probability of misclassification followed by city block distance.