Data matrices with rows which are independent samples from some population were considered by Fisher and Wishart in the 1930?s. More, recently the abstract sets of such random matrices have been studied at length and their limit theorems proven. Some of these results along with the associated entropies will be discussed and their applications in modern communication theory will be mentioned. Keywords: Random matrices, Free probability, Entropy.

   N.A

  Rural and agricultural marketing involve, in its simplest form, buying and selling of rural and agricultural produces. In the process of development of marketing, many intermediaries also emerged between the farmers and the consumers which had consequently reduced the farmers? share of the price paid by the consumers. In this direction, the regulated markets were established to regulate effectively the sale and purchase of agricultural produce and establish market yards for marketing of agricultural produce so as to free the farmers from the clutches of middlemen. Keywords: Regulated market, Market efficiency, Data envelopment analysis.

  N.A

  N.A

  The level of development of different districts of West Bengal was obtained with the help of composite index based on optimum combination of forty five developmental indicators. Eighteen districts of the State have been included and the data on various indicators for the year 2001-02 were used in the study. The level of development was examined separately for agricultural sector, infrastructural facilities and socio-economic sector. Wide disparities in the level of development are observed among different districts. Infrastructural facilities and agricultural development are found to be positively associated with socio- economic development. For bringing out uniform regional development, potential targets of important indicators have been estimated. Keywords: Composite index, Developmental indicators, Model districts, Potential target.

  N.A

  N.A

  N.A

  The Richards curve is frequently used to model cumulative growth in science. Recent studies have shown that the resulting residuals in regression applications fitting the Richards curve are autocorrelated. Such autocorrelation has typically not been taken into account in the regression model, which calls into question the subsequent statistical inferences based on the model. This paper shows how fitting the derivative function of the Richards curve to the incremental weight gains, i.e. to the ?first differences?, may mitigate autocorrelation inherent in growth data. The procedure is illustrated on turkey growth data from Syria. This new approach should apply to growth data in general, and to suitable poultry growth data in particular. Keywords: Logistic growth model, Power-law logistic model, Durbin-Watson test, First differences.

  The National Sample Survey Organisation (NSSO) surveys are the main source of official statistics in India. A range of invaluable data at the macro level (e.g. state and national level) is generated through these surveys. However, the NSSO data cannot be used directly to produce reliable estimates at the micro level (e.g. district or further disaggregate level) due to small sample sizes. There is a rapidly growing demand of such micro level statistics in India as the country is moving from centralized to more decentralized planning system. In this article we employ small area estimation (SAE) techniques to derive model- based estimates of proportion of poor households at district level in the State of Uttar Pradesh in India by linking data from the Household Consumer Expenditure Survey 2006-07 of NSSO 63rd round and the Population Census. The poverty line used in this study is same as those of year 2004-05, given by Planning Commission, Govt of India. The poverty line is used to identify whether a given household is poor or not. A household having monthly per capita consumer expenditure below the state?s poverty line is categorised as poor household. The results show that the model-based estimates are reliable. In contrast, the direct estimates are very unstable. These estimates are expected to provide invaluable information to policy-analysts and decision-makers. Keywords: NSSO survey, Population census, Small area estimation, Poor households, Proportion.

  Optimal design of multi-response mixture experiments for estimating the parameters of multi-response linear models is an interesting problem which is yet to be discussed in literature. In this paper, we present a formulation of optimal design for multi-response linear models using Semi-Definite Programming (SDP) that can generate D-, A- and E-optimal designs. We generate both approximate and n-exact D-optimal designs for multi-response linear models for multi-response mixture experiments. The proposed method has an advantage as using it one can use SDP solver of TOMLAB (Holmstrom 2004) and YALMIP (Löfberg 2005) software within MATLAB environment in order to minimize the amount of computation time needed to generate an optimal design for multi-response models. Keywords: D-optimal design, Multi-response linear model, Multi-response mixture experiment, Semi-Definite programming.

  Agriculture is an occupation with risks, especially, when farmers use more machines for farming. Therefore, it is important to associate risk factors with agricultural related injuries, which has implications to make policies for preventing injuries. Lyman, et al. (1999) and others have applied multiple logistic regression in agricultural research. However, it is challenging to incorporate appropriate interactions in the logistic regression model. In this paper, the approach of classification tree is used to associate risk factors with farm injuries, which can automatically handle possible interactions. Based on a classification tree, it is also possible to develop different prevention programs for various subgroups given by the tree. We used a data set of 1,051 subjects from the states of Alabama and Mississippi, USA, including African-American farm workers and Caucasian and African- American owners (Source: UAB Injury Center). Condition of farm machinery (Excellent/Good vs. Fair/Poor) first splits the tree. For fair/poor condition of farm machinery, large farms (749 or more acres) give high injury rates (0.28), while for farms with fewer than 749 acres, the estimated injury rate is 0.16 without farm safety training and it is reduced to 0.063 with farm safety training. For excellent/good condition of farm machinery, history of farm injury then splits the tree. Factors that further split the tree include ?How Often Do I Hurry When Doing Farm Work on Machinery? wearing a seat belt, education, primary commodity, race, alcohol consumed per week and farm safety training. Furthermore, classification trees are developed separately for Caucasians and African-Americans. Keywords: Machines, Farm injury, Socio-demographic characters, Risk factors, Comorbid medical conditions, Nonparametric approach, S-PLUS, Classification trees.

  Personnel Management Information System Network (PERMISnet- II) has been designed for Indian Council of Agricultural Research (ICAR) using .NET technology. System is hosted at Indian Agricultural Statistics Research Institute (IASRI) web server and can be accessed through web address http://permisnet/iasri.res.in, using any web browser. System has been designed using modular approach and has separate modules for managers, general user, ICAR personnel and nodal officers of different Institutes. Nodal officers have the right to insert, edit/update and delete data related to their respective Institutes. Research Manager Personnel (RMP?s) of ICAR have the flexibility to view reports at all levels, i.e. Institutional, Subject Matter Division (SMD) and consolidated for entire ICAR on different parameters. ICAR personnel have the right to view and download their bio-data. Keywords: Personnel Management Information System, PERMISnet, PERMISnet II, ICAR, Manpower.