In this paper, a transformed class of estimators has been developed for estimating the mean of a finite population under two-phase sampling usingtwo auxiliary variables. The mathematical expressions for bias and mean square error (MSE) of the proposed class, as well as for the pre-existingestimators, have been derived to the first order of approximation. The proposed class of estimators has been compared with the other well-knownestimators using the MSE criterion. Moreover, the optimum sample sizes of the first-phase and second-phase samples, along with the optimumMSEs of the concerned estimators, have been derived using the cost function analysis. The theoretical results have been empirically validated byconsidering real population datasets.Keywords: Auxiliary variable, Bias, Mean square error, Percent absolute relative bias, Percent relative efficiency, Study variable, Two-phase sampling.

  In the present paper we have proposed dual to ratio-cum-product type estimator for estimation of the population mean in double sampling for stratification. The motivation of proposed estimator is based on Singh (1967) and Lone et al. (2020). We have derived expressions for bias and MSE for the proposed estimator. The mean square error of the proposed estimator is compared with usual unbiased estimator of population mean in double sampling for stratification, Ige and Tripathi (1987) estimators, Tailor et al. (2015) estimator and Lone et al. (2020) estimators. We have obtained the conditions under which proposed estimator is more efficient than other estimators. The paper concludes with a numerical illustration.

  Time-series data are rarely purely linear or nonlinear and often contain both these patterns. In this article, hybrid models are developed by combining linear Kalman Filter (KF) and Nonlinear Least Squares Support Vector Machine (LSSVM) methodologies for time-series forecasting. Particle Swarm Optimization (PSO), which is a very efficient population-based stochastic optimization technique is employed to estimate the hyper-parameters of these models. The relevant computer program is written in MATLAB function (m file) and MATLAB software package is used for data analysis. As an illustration, developed hybrid models are applied to all-India monthly rainfall time-series data. The superiority of these models over individual linear KF and nonlinear LSSVM methodologies is demonstrated for the data under consideration using Root Mean Square Error (RMSE) and Mean Absolute Percent Error (MAPE) criteria.

  Generalized Row-Column (GRC) designs are defined as designs with v treatments in p rows and q columns such that the intersection of each row and column (cell) consists of k experimental units. In GRC designs, since there are more than one number of units in a cell, it is likely that the treatment applied to one experimental unit may affect the response of the neighbouring unit in the same cell if the units are placed linearly adjacent giving rise to spatial effects. The study in presence of spatial effects from neighbouring units requires construction of an arrangement in which the neighbouring units have to appear in a predetermined pattern. Here, series of GRC designs balanced for these spatial effects have been developed. The information matrices for estimating the contrasts pertaining to direct effect and spatial effect have been derived. The designs developed ensure that within a cell every treatment has every other treatment appearing as neighbour a constant number of times. A list of efficient designs has been prepared. Further, in order to give a readymade solution to the experimenters, a SAS macro has been developed that generates the layout of the designs with parameter v (prime), p = v, q = v-1, k = s (3 ≤ s ≤ v-1).

  Accurate and timely price information and forecasting help in making efficient plans and strategies. Non-linearity and non-stationarity behaviour of price data create problems in price forecasting. In this paper, variational mode decomposition (VMD) based optimised genetic algorithm (GA) hybrid machine learning (ML) models have been proposed. The VMD algorithm is employed to decompose the price data into intrinsic mode functions (IMFs) which is further forecasted using ML models namely support vector regression (SVR) and random forest (RF). The practical use of the SVR and RF models is limited because the accuracy of ML models heavily depends on a proper setting of hyper-parameters. Therefore, these model hyper-parameters are optimized using GA. Further, the forecasted values of IMFs through the GA optimised SVR and RF are aggregated for the final forecast. The results of the proposed model are benchmarked with the comparative models. The proposed VMD-GA-RF and VMD-GA-SVR models are tested on the weekly onion price of the Delhi and Nashik market. The results clearly demonstrate that the combination of VMD and GA optimized models can improve the performance of the prediction of the dataset.

  Sampling techniques play an important role in determining the efficiency of control charts. The study was designed to develop a new Shewhart-type x control chart to monitor processes using a newly developed cost-effective method of ranked set sampling namely Stratified Balanced Quartile Ranked Set Sampling (SBGQRSS). SBGQRSS is a recent sampling design proposed based on the traditional sampling of ranking set samples. ARL is utilized as a performance measure to evaluate the efficiency of SBGQRSS x control chart and other considered SRS, ranked set sampling (RSS) and extreme ranked set sampling (ERSS) charts by using Monte Carlo simulations. In most simulation scenarios, the SBGQRSS control chart is the best in comparison to RSS and its extensions.An application to real forest data illustrates the proposed method, with an evident increase in the sensitivity of the SBGQRSS based Shewhart-type x chart compared to other control charts.

  In the Global Hunger Index 2021, India ranks 101st among the 116 countries, with almost one-third of all undernourished children living in the country. Instability in production, market and other risks make agriculture a riskier enterprise, affecting farmers? income and therefore food security. So, the Pradhan Mantri Fasal Bima Yojana (PMFBY) was introduced by the government to alleviate crop uncertainties in the interest of the country?s farmers and to safeguard them from the risky nature of farming, as part of the ?One Nation-One Plan. The findings of this study highlight that overall, 51.16 per cent of beneficiaries and 13.16 per cent non-beneficiaries were significantly aware of PMFBY. The least per cent of beneficiaries and non- beneficiaries were aware of the risk covered under the scheme (51.64%) and procedure for insuring crops (14.97%) respectively. Regarding the overall awareness level towards the PMFBY, it was found that among the beneficiaries, maximum respondents were significantly aware (51.10%), followed by unaware (29.99%) and then moderately aware (18.88%). On the contrary, it was reported that the maximum number of non ? beneficiaries were unaware (64.93%), followed by moderately aware (31.18%) and then significantly aware (13.26 %).PLUM (Polytomous Universal Model) method was used to analyze the relationship between perceptions of farmers and adoption of PMFBY. It was observed that there exists significant relationships, for both beneficiaries and non ? beneficiaries. The parameter estimates showed that PMFBY acts as ?a safeguard against production losses? and it has a significant, positive and highest impact (101.39 times) on the adoption of the scheme among beneficiaries and perception ?Farmers? friendly procedure in buying crop insurance? has shown a significant and most positive effect (28.70 times) on adoption in non-beneficiaries. Such crop insurance schemes are way more beneficial to the farmers to prevent them from huge losses. So there is an urgent need to make people aware of the scheme and help them take the benefits by spreading awareness among them. Farmers should be made aware Regarding the agencies involved, crops and risks covered, the procedure of insuring crops, premium to be paid, source of the required information and so on in order to make PMFBY a success.

  Neural network models offer a number of advantages as they have an ability to tactically detect complex non linear relationships between dendrometric variables of tree, which are very helpful in tree height modeling. In this study, artificial neural network (ANN) models and nine conventional height diameter equations were employed to validate the height diameter relationship in Chir Pine plantations. Height diameter measurements of 1500 Chir Pine trees in150 sample plots from three forest divisions of Jammu province of UT of J&K, India were used. For the purpose of developing and validating models, the data was randomly partitioned into training (80%) and testing (20%) sets. All the fitted height diameter models resulted in significant coefficients, which indicated that these models were able to capture the underlying height diameter relationships. Out of the nine traditional height diameter models, M7 height diameter model had the highest fitting precision, with lower values for Akaike Information criteria (AIC),Bayesian information criteria (BIC). However, under cross validation artificial neural networks (ANN) outperformed conventional models in every aspect as they resulted in lower values of prediction error rates (PER) and other selection criteria, where neural network model with 10 numbers of neurons came out be superior in comparison to other fitted models.

  There exist several publication recommendation systems which recommend publications to researchers and academicians based on their area of interest. A publication recommendation system specifically meant for recommending the agricultural publications will help the agricultural researchers and academicians in getting recommendations in the form of new publications published in their area of interest/work. The developed system uses the keywords and title of the publications to find out the similarity between the newly added publication and all existing publications in the publication repository. The TF-IDF model and the Cosine similarity measure have been used for finding out the similarity using R statistical software. The recommendation of the new publication is sent to the authors of the five most similar publications through mail. The system has been tested using the data from KRISHI Publication and Data Inventory Repository. The agricultural publication recommendation system successfully sent publication alerts to the authors of most similar publications from the data. This system can help the agricultural researchers by recommending the newly published research to the authors of existing publications working in the similar area using an email alert. It will support the researchers to stay up-date with the latest work carried out in their area of interest.

  Vitamins are a diverse group of primary metabolites. These are produced in modest amounts, making it challenging to research the related pathways and enzymes. The development of genetic sequence information and the need to boost the nutritional value of plants by boosting their vitamin content have both substantially aided in the analysis of vitamin production in plants at the molecular level. Here, we have compared four most popular supervised machine learning algorithms: Support vector machine (SVM), Naive Bayes (NB), Random Forest (RF) and K-nearest neighbor (KNN) for classifying the vitamin biosynthesis genes. We first carried out binary classification to classify genes as vitamin biosynthesis related and not related. Further, vitamin biosynthesis genes were classified into 10 vitamin classes (Vit A, Vit B1, Vit B2, Vit B5, Vit B6, Vit B7, Vit B9, Vit C, Vit E, and Vit K). Our results for binary classification suggested Random forest to be the best classifier based on various evaluation parameters including accuracy, precision, sensitivity, specificity, F1 score and AUC (Area under Curve of ROC (Receiver Operating Characteristic curve)). Whereas, for multiclass classification of vitamin biosynthesis genes, KNN was found to be the best classifier on the basis of Accuracy, Matthews correlation coefficient (MCC) and Area Under ROC curve (AUC).

  Hindi Summaries of papers

  Acknowledgement to Reviewers