Pre-harvest Forecast Research Articles

Different statistical models based on weather parameters in Navsari district of Gujarat

Agriculture plays very important role in development of country. Rice is a staple food for more than half of world’s population. Timely and reliable forecasting provides vital and appropriate input, foresight and informed planning. The present investigation was carried out to forecast Kharif rice yield using two different statistical techniques, viz., discriminant function analysis and logistic regression analysis. The statistical models were developed using data from 1990 to 2012 and validation of developed models was done by using remaining data, i.e., 2013 to 2016. It was observed that value of adjusted R2 varied from 73.00 per cent to 93.30 per cent in different models. The best forecast model was selected based on high value of adjusted R2, Forecast error and RMSE. Based on obtained results in Navsari district, the discriminant function analysis technique (Model-5) was found better than logistic regression analysis (Model-12) for pre-harvest forecasting of rice crop yield. The results revealed that Model-5 showed comparatively low forecast error (%) along with highest value of Adj. R2 (93.30) and lowest value of RMSE (120.07). Also Model-5 is able to generate yield forecast a week earlier (39thSMW) than Model-12 (40thSMW).

Development of yield forecast models with raw data through stepwise regression analysis (forward method) and discriminant function analysis for pigeon pea crop on Chhattisgarh plain zone

Under this investigation we studied about minimization of weekly data through stratification and Simple Randon Sample technique. Individual effect of weather parameters has been considered under the study. Forecasting model developed on two statistical technique i.e. Stepwise Regression Analysis (Forward Method) technique and Discriminat Function Analysis technique. Both the techniques are found very suitable for the district wise and zone level pre-harvest forecast for the pigeonpea crop on Chhattisgarh. Models fitted with the Stepwise Regression Analysis on 35 variables and Time Trend (T) for Zone. The developed model is showing best fit on the basis of very high value of significance and maximum R2 (84%) for the plain Zone, highly significant at 0.1% level of significance and models fitted with DFA then 2 Discriminant score (ds1 and ds2) and Time Trend (T) the models found highly significant and R2 value 69% for plain zone for Pigeonpea crop. The developed model has been validated by the error parameters viz. Minimum MAE, Minimum MSE, Minimum RMSE, Minimum PE and Minimum PD along with maximum R2. These developed models are useful to farmers of Chhattisgarh to decide their future prospects and possible course of action in advance. This was very challenging task for the researchers to develop a precise & accurate and best fit model for the future forecast. May this study will clear the fog through new methodology and also will create the new way on the direction of forecast.

Pre-harvest forecast of kharif rice yield using PCA and MLR technique in Navsari district of Gujarat

In this paper Principal Components (PC) and Multiple Linear Regression (MLR) Technique were used for development of pre-harvest model for rice yield in the Navsari district of south Gujarat. The weather indices were developed and utilized for development of pre-harvest forecast models. The data of rice yield and weather parameters from 1990 to 2012 were utilized. The cross validation of the developed forecast model were confirmed using data of the years 2013 to 2016. It was observed that value of Adj. R2 varied from 89 to 96. The appropriate forecast model was selected based on high value of Adj. R2. Based on the outcomes in Navsari district, MLR techniques found to be better than PCA for pre harvest forecasting of rice crop yield. The Model-2 found competent to forecast rice yield in Navsari district before eight weeks of actual harvest of crop (37th SMW) i.e during reproductive stage of the crop growth period.

Weather relation of rice-grass pea crop sequence in Indian Sundarbans


 In order to develop weather-based yield prediction models for rice and grass pea in coastal saline zone of West Bengal, the experiments were conducted with rice (cv. CR 1017) and grass pea (cv. Bio L 212) in the rainy and winter seasons, respectively of 2016-17 and 2017-18. Rice was sown in nursery bed on six different dates starting from June 15 to July 19 at weekly interval in both rainy seasons in two different land situations viz. medium upland and medium lowland. Likewise, grass pea was sown on six different dates just before harvesting of rice. It was observed that both early sown rice and grass pea resulted in higher grain yield and took more time to mature under medium lowland situation irrespective of sowing dates. Correlation study revealed that air temperature during sowing to transplanting phase exhibited significant positive correlation with grain of rice in medium upland (Tmax = 0.76**, Tmin = 0.69*) and medium lowland (Tmax = 0.93**, Tmin = 0.81**) situations. On the other hand, maximum temperature and total solar radiation during 100% emergence to 100% flowering stage were negatively associated with the grain yield of grass pea in both medium upland (Tmax = -0.69*, Accumulated solar radiation = -0.73**) and medium lowland (Tmax = -0.74**, Acc. solar radiation = -0.77**) situations. Grain yield of rice and grass pea could be predicted with 94.4% and 87.4% predictability. Pre-harvest forecasting of grain yield was possible with 77.3% for rice and 83.8% for grass pea.

Pre-harvest forecast of rice yield based on meteorological parameters using discriminant function analysis

The present study is focused to develop pre-harvest forecast models for rice yield based on meteorological parameters in Haryana. The discriminant function analysis (DFA) technique has been used for forecasting the rice yield. For this study, fortnightly weather data from 1980–81 to 2018–19 on five weather variables namely maximum temperature, minimum temperature, average relative humidity, sunshine hours and accumulated rainfall have been used. The data for period from 1980–81 to 2014-15 have been utilized for model building and subsequent four years data for the period 2015–16 to 2018-19 have been used for model validation. The residuals obtained after fitting regression model by taking rice yield (dependent variable) and year (independent variable) using data for the period 1980–2015 have been used to categorize rice yield into two/three groups. Various performance measures have been used to measure the performance of the developed models at different fortnights. The results based on various performance measures indicated that 21st fortnight (29 September–13 October) or one month before harvest is the best time for forecasting the rice yield. The findings of the present study may be helpful for policy planners and various stakeholders to take appropriate decisions to make arrangement for procurement, distribution, storage, trade in domestic and international markets as well as for managing the proper inventory in advance.

Pre Harvest Forecasting of Kharif Rice Yield Using Weather Parameters for Strategic Decision Making in Agriculture

In the recent year, pre harvest crop yield forecasting has been a topic of interest for producers, policy makers, government and agricultural related organizations. Pre harvest crop forecasting is important for national food security. Construction of appropriate yield forecast promotes the output of scenario analyses of crop production at a farm level, which enables suitable tactical and strategic decision making by the farmer. Indeed, considerable benefits apply when seasonal forecasting of crop performance is applied across the whole value chain in crop production. Timely and accurate yield forecast is essential for crop production, marketing, storage and transportation decisions as well as for managing the risk associated with these activities. In present manuscript efforts were made for development of pre harvest forecast models by using different statistical approaches viz. multiple linear regression (MLR), discriminant function analysis and ordinal logistic regression. The study utilized the crop yield data and corresponding weekly weather data of last 30 years (1985-2014). The model development was carried out at 35th and 36th SMW (Standard Meteorological Week) for getting forecast well in advance of actual harvesting of the field crop. The study revealed that method of discriminant function analysis gave best pre harvest forecast as compare to remaining developed models. It was observed high value of Adj. R2= 0.94, low value of RMSE= 164.24 and MAPE= 5.30. The model can be used in different crop for reliable and dependable forecast and these forecasts have significant value in agricultural planning and policy making.

Development of State Space Models with Weather as Exogenous Input for Sugarcane Yield Prediction in Haryana

Parameter constancy is a fundamental issue for empirical models to be useful for forecasting, analyzing or testing any theory. Unlike classical regression analysis, the state space models (SSM) are time varying parameters models as they allow for known changes in the structure of the system over time and provide a flexible class of dynamic and structural time series models. The work deals with the development of state space models with weather as exogenous input for sugarcane yield prediction in Ambala and Karnal districts of Haryana. The state space models with weather as exogenous input using different types of growth trends viz., polynomial splines; PS(1), PS(2) and PS(3) have been developed however PS(2) with weather input was selected as the best suited model for this empirical study. Timely and effective pre-harvest forecast of crop yield helps in advance planning, formulation and implementation of policies related to the crop procurement, price structure, distribution and import-export decisions etc. These forecasts are also useful to farmers to decide in advance their future prospects and course of action. The sugarcane yield forecasts based on state space models with weather input showed good agreement with state Department of Agriculture and Farmers’ Welfare yield(s) by showing nearly 4 percent average absolute relative deviations in the two districts.

English

The study was carried out at Department of Agricultural Meteorology, VNMKV, Parbhani to find the quantitative relationship between weather parameters and district level yield of pigeon pea. For this purpose, 30 years (1982-83 to 2011-12) weather and crop yield records of different district of Marathwada region of Maharashtra (India) were collected. A twenty-eight-week crop period model was recommended for pre harvest forecast due to higher R2 value and lower simulated forecast deviation. The time trend, maximum temperature, morning and evening relative humidity significantly affected crop yield. In Aurangabad equation shows that the Tmin. and Tmean has positively significant effect on yield. In latur the equation shows that all weather parameters have highly positively significant effect Tmax., Tmin., Tmean, RH I, RH II, Rainfall and R.D. effect on yield.In osmanabad equation show that the Tmin.and Tmean have negatively significant effect on yield while other variables show non-significant effect on yield. While in Parbhani and Beed District all the weather parameters have no significant effect on yield.

English

English

Pre harvest forecasting of crop yield using non-linear regression modelling: A concept

The concept of pre-harvesting of crop yield using nonlinear growth models and detrended yield for developing yield forecast model is rarely employed in forecasting. A novel approach attempted in this study to use nonlinear models with different weather variables and their indices and compare them to identify a suitable forecasting model. Weather indices based regression models were developed using weather indices as independent variables whiledetrended yield (residuals) was considered as dependent variable. The approach provided reliable yield forecastabout two months before harvest.

An application of principal component analysis for pre- harvest forecast model for wheat crop based on biometrical characters

An application of principal component analysis for the development of suitable statistical models for pre-harvest forecast of wheat yield based on biometrical characters has been illustrated in the present paper. The data obtained from the experiments on wheat under normal and late sowing situations have been utilised to develop the model. The result have revealed that the proposed model can provide reliable pre- harvest forecast of wheat yield in both the situations within the reasonable range of per cent standard error of 2.16 to 4.96 per cent.

Wheat yield forecast using detrended yield over a sub-humid climatic environment in five districts of Uttar Pradesh, India

A study was carried out to forecast the yield of the wheat crop for five districts of Uttar Pradesh namely Lucknow, Kanpur, Banda, Jhansi and Faizabad. The daily weather data on variables such as maximum temperature, rainfall, minimum temperature, and relative humidity were arranged week wise from sowing to harvesting and the relations between the weather variables and yield was worked out using statistical tools like correlation and regression. The yield has been detrended by obtaining the parameter estimates of the model and subsequently the detrended yield was used to forecast the yield of the crop using ARIMA model. The proposed method of obtaining pre-harvest forecasting of yield of crops was compared with the traditional approaches of forecasting and the proposed method was evaluated in terms of criteria's such as goodness of fit of the model. It was observed that in all the districts the proposed model performed better as compared to the traditional method both in terms of goodness of fit as well as forecasting performance. Thus it can be concluded that the proposed approach is better and more suitable as compared to the traditional approach for forecasting the wheat yield in the five districts of Uttar Pradesh.

An application of principal component analysis for pre- harvest forecast model for rice crop based on biometrical characters

An application of principal component analysis for the development of suitable statistical models for preharvest forecast of rice yield based on biometrical characters has been dealt with in the present paper. The data obtained from the two experiments on rice have been utilised to develop the model. The forecast yields of based onthese models have been found to be 24.25, 22.60 and 21.10 q/ha against the actual yield of 28.00, 23.56 and 21.85 q/ha, respectively, in experiment –I. For experiment –II the forecast yields were found to be 24.62, 28.06 and 29.43 q/ha against the actual yield of 28.82, 29.31 and 26.59 q/ha, respectively. These forecast yields are subject to maximum of almost 10 percent standard error. In most of the cases, the forecast yields were found to be close to the actual yield except in some cases. The values of R2, i.e. 79.80 and 72.60 for experiment –I and II, respectively, indicate the validity of the models. Statistical tool like viz. principal component analysis (PCA) has been first time applied to develop pre-harvest forecast model based on experimental data.

Pre-harvest forecast of pigeon-pea yield using discriminant function analysis of weather variables

In the present paper, an application of discriminant function analysis of weather variables (minimum & maximum temperature, Rainfall, Rainy days, Relative humidity 7 hr & 14 hr, Sunshine hour and Wind velocity )for developing suitable statistical models to forecast pigeon-pea yield in Faizabad district of Eastern Uttar Pradesh has been demonstrated. Time series data on pigeon-pea yield for 22 years (1990-91 to 2011-12) have been divided into three groups, viz., congenial, normal, and adverse based on de-trended yield distribution. Considering these groups as three populations, discriminant function analysis using weekly data on eight weather variables in different forms has been carried out. The sets of discriminant scores obtained from such analysis have been used as regressor variables along with time trend variable and pigeon-pea yield as regressand in development of statistical models. In all nine models have been developed. The forecast yield of pigeon-pea have been obtained from these models for the year 2009-10, 2010-11 and 2011-12, which were not included in the development of the models. The model 4 and 9 have been found to be most appropriate on the basis of R2adj, percent deviation of forecast, percent root mean square error (%RMSE) and percent standard error (PSE) for the reliable forecast of pigeon-pea yield about two and half months before the crop harvest.

Yield Forecasting Models based on Weather Parameters for Eastern U.P.

Yield Forecasting Models based on Weather Parameters for Eastern U.P. In the present study, attempts have been made to development of pre harvest models for pre harvesting forecasting of rice yield at district level on the basis of generated weather variables. Weekly data (14 meteorological weeks) because Time of the pre harvest forecast is 14th week 2½ month before harvesting flowering stage and seven weather variables over a span of 20 years period (1995 to 2014) has been used along with the annual rice production data for respective year for eastern UP. The data of de-trend yield and generated weather variables for 18 years of has been used for generation of the models. (1995-2012). Total five models were validated with 2 year independent data set (2013 and 2014). The (PPE) ranging - 4.32 to 10.56 % from the observed yield for all 5 models. Highest correlation has been found 0.68 between Z231 (Weighted interaction between MaxT and MinT). Significance of models are obtained by on the basis of highest R2 (86%) and by (P-value) (0.08). MBE (Mean Biased Error), RMSE (Root Mean Square Error) and PE (Percent Error) has been used for the Error analysis. On the basis of R2 and other parameter the model IV has been found best..

Different approaches on pre harvest forecasting of wheat yield

Agriculture is backbone of Indian economy, contributing about 40 per cent towards the Gross National Product and provide livelihood to about 70 per cent of the population. According to the national income published in Economic survey 2014-15, by the CSO, the share of agriculture in total GDP is 18 percent in 2013-14. The Rabi crops data released by the Directorate of Economics and Statistics recently indicates that the total area coverage has declined; area under wheat has gone down by 2.9 per cent. Therefore needs to be do research to study weathersituation and effect on crop production. Pre harvest forecasting is true essence, is a branch of anticipatory sciences used for identifying and foretelling alternative feasible future. Crop yield forecast provided useful information to farmers, marketers, government agencies and other agencies. In this paper Multiple Linear Regression (MLR) Technique and discriminant function analysis were derived for estimating wheat productivity for the district of Varanasi in eastern Uttar Pradesh. The value of Adj. R2 varied from 0.63 to 0.94 in different models. It is observed that high value of Adj. R2 in the Model-2 which indicated that it is appropriate forecast model than other models, also the value of RMSE varied from minimum 1.17 to maximum 2.47. The study revealed that MLR techniques with incorporating technical and statistical indicators (Model 2) was found to be better for forecasting of wheat crop yield on the basis of both Adjusted R2 and RMSE values.

Pre-harvest forecast models for wheat yield based on biometrical characters

Some statistical models for pre-harvest forecast of wheat yield based on biometrical characters in situation of normal and late sowing of wheat have been developed in the present paper. In both the situations, linear multiple regression model (model-I), where biometrical characters are used in the original form, has been found to be the best forecasting model as it has consistently smaller percent standard errors for the forecast yield of wheat alongwith maximum value of R2adj. (49 to 51%).

Pre-harvest forecasting of county wheat yield and wheat quality using weather information

County wheat yield and wheat quality are forecast using weather information. Regression models are estimated to account for the effect of weather on county wheat yield, protein, and test weight. The explanatory variables include precipitation and temperature for growing periods that correspond to biological wheat development stages. Wheat yield, protein, and test weight are strongly influenced by weather. The forecasting power of the yield and protein models is enhanced by adding a spatial lag effect. Out of sample forecasting tests confirm the models’ usefulness in predicting wheat yield and wheat quality.

Estimation of cotton yield based on weather parameters of Banaskantha district in Gujarat state

The present investigation was undertaken to identify the quantitative relationship between weather parameters and district level yield of cotton and to develop preharvest forecast models for cotton yield. For this purpose 32 years weather and crop yield records of Banaskantha district were collected. It was found that the 26 week crop period model ( using original weather variables, week wise approach) was recommended for pre harvest forecast due to higher R2 value and lower simulated forecast error. The time trend, maximum temperature, morning and evening relative humidity have significantly affected on crop yield.

Estimation of cotton yield based on weather parameters of Junagadh district in Gujarat state

The study was carried out to find the quantitative relationship between weather parameters and district level yield of cotton. For this purpose 32 years weather and crop yield records of Junagadh district (India) were collected. A twenty six week crop period model was recommended for pre harvest forecast due to higher R2 value and lower simulated forecast deviation. The time trend, maximum temperature, morning and evening relative humidity significantly affected crop yield.