SOFTWARE FOR FITTING OF DISTRIBUTIONS
KrishiKosh
View Archive InfoField | Value | |
Title |
SOFTWARE FOR FITTING OF DISTRIBUTIONS
|
|
Creator |
ASHUTOSH KARNA
|
|
Contributor |
R. C. Goyal
|
|
Subject |
marketing, electrophoresis, sampling, physical control, land resources, sets, packaging, manpower, byproducts, diseases
|
|
Description |
t-8154
Statistics is a science of drawing conclusions on uncertainty and deals with collection, presentation, analysis and interpretation of numerical data. One of the main objectives of statistics is to draw valid inferences about a population from the analysis of a sample drawn from that population. It generates the need to study probability distribution and analyzing the behavior of data distribution in order to estimate various statistical constants. Probability distributions fall in two major categories: 1) Discrete distribution-when a random variable is allowed to take on only a limited number of values, e.g. Uniform, Binomial, Poisson, Geometric etc. 2) Continuous distribution-When a random variable can take any value within a specified range e.g. Normal, Beta, and Exponential etc. The theoretical probability distributions provide decision makers with a logical basis for making decisions on the basis of limited information. For example, the distribution of weeds over a unit area in a field crop is fitted by Uniform distribution; but distribution of insects in a field is fitted by Poisson distribution. Similarly, Logistic and Pareto distribution are used to fit population data. Thus, one must be careful while choosing the correct probability distribution as per the situation being faced. The theory of estimation forms the foundation of statistical inference which uses the probability concepts to deal with uncertainty and decision making. The concept of estimation of parameters can’t be visualized until and unless the distribution function of population is identified and thus distribution fitting also plays a significant role in finding various statistical constants of the population. Thus an attempt has been made to develop software that can fit certain number of discrete and continuous distributions as per the real life situations and fit the theoretical data using Kolmogorov - Smirnov and Chi-Square tests of goodness of fits, once the parameter have been estimated (broadly using MLE techniques). Leading statistical softwares including SAS, SPSS, and Minitab etc. are already available on this aspect yet no indigenous and/or web-enabled software has thus far been developed at IASRI. Keeping this in view, this software has been developed using Visual C# 2008 language with ASP.NET 3.5 framework. Data file formats supported by this software are .txt and .xls. In addition to fitting of distribution, the software can also be used interactively to draw Quantile-Quantile plot for continuous distributions; and computing various special functions (including Gamma, Beta, and Incomplete Gamma etc.) at runtime. A user friendly help file with step-by-step method for using the software has also been provided. |
|
Date |
2016-12-23T09:27:35Z
2016-12-23T09:27:35Z 2009 |
|
Type |
Thesis
|
|
Identifier |
http://krishikosh.egranth.ac.in/handle/1/92326
|
|
Format |
application/pdf
|
|
Publisher |
INDIAN AGRICULTURAL STATISTICS RESEARCH INSTITUTE
|
|