Modeling of Lemon Production: A Stepwise Regression Approach

Abstract

Lemon, which is both tropical and subtropical fruit, was formerly used for only entertaining the guests, but today, as a result of its many different uses; it is highly popular with both consumers and producers. Bangladesh is home to extensive lemon groves due to the fruit's high nutritional value, favorable growing conditions, and widespread demand on both domestic and international markets. This study examined the combined impact of farm specific factor and climatic variable on lemon production in Bangladesh. Here the stepwise regression technique with various diagnostic plots in R studio was used to assess the suitable multiple linear regression model and an effective autoregressive integrated moving average (ARIMA) model-based time series analysis was produced for future projection of lemon. The study found a positive link between lemon output and price but a negative correlation with climate variables. Annually, one substantial drop in the lemon cultivars as a result of climatic stress is initiated due to the most significant aspects of which are temperatures and rainfall. Result found, with each millimeter of increased AMR, output diminished by 0.0006419 times on average. The production drops by 0.3564144 times on average for every unit increase in annual mean temperature (AMT). Price increases for lemons may be traced back to covid-19, when production was increased in response to increasing demand about the health benefits of lemons. Also, the investigation found ARIMA (1,1,1) model with drift to the best for future projections of lemon production.