• A. Lagat Department of Basic and Applied Science, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya
  • C. Biwott ASA Kenya Limited, Nairobi, Kenya
  • R. Komen Equity Bank Limited, Nairobi, Kenya
Keywords: time series analysis, ARIMA models, price fluctuations, food production, agriculture


Agriculture is considered a fundamental sector employing over forty percent of the population in Kenya and contributes a large proportion of about 26% of the Gross Domestic Product. To improve investment in this vital sector and ensure secured food production to the increasing population, it is crucial that farmers are aware of the best market opportunities that maximizes consistent income. This work focused on determining the volatility of farm produce prices as a way of establishing opportunities of farming investment.
The trends and volatility analysis used average monthly prices of agricultural produce collected and published by the Ministry of Agriculture Livestock and Fisheries (MoALF). It covered the period January 2013 – July 2018The data extracted was restricted to Maize, Beans, Cowpeas and potato crops. Various ARIMA (p,d,q) time series models were fitted for each respective crop and respective best model was selected based on Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) statistics. The analysis were done using R software (V3.4.2) and all tests were carried at 5% level of significance.
The analysis identified and fitted unique models for the Irish potato, maize, beans and cowpea to be ARIMA (2,0,2), ARIMA (2,1,0), ARIMA (0,1,1) and ARIMA (1,1,1) respectively. It would be appropriate to use high frequency data, determine the spatial distribution of the prices over the years as well as the effect of co-integrating oil prices the models.