Time Series Arima Modelling Of Loan Borrowing And Repayment At Kifanya Saccos Limited In Njombe, Tanzania

ABSTRACT

Savings and credit cooperative societies (SACCOS) in developing countries like Tanzania

are very importance in solving the financial problem to low income groups especially in

rural areas. SACCOS have the sole role of mobilizing savings that creates the source of

funds to benefit their members. But for a SACCOS to prosper in its role of financial

intermediation, it has to give more loans and ensure that most of them are repaid in time.

To plan for such achievements the SACCOS needs accurate forecasting techniques to

predict the future value of its loan portfolio.

This study was conducted at Kifanya SACCOS in Njombe Region, Tanzania. The study

focuses on modelling and forecasting of the monthly loan borrowing and repayment on

short-term basis by using the autoregressive integrated moving average (ARIMA) models.

The modelling process was preceded by analysing the components of the series of interest

which revealed the presence of increasing trend and cyclic variations. Seasonal variation

appeared to be very significant in loan borrowing only.

The findings showed that the best time series model for monthly loan borrowing is

ARIMA (0,1,1)(0,0,1)12 whilst that for loan repayment is ARIMA (1,1,2). Also, it was

revealed that the developed models fit well the historical data and can be used in shortterm

forecasting of the future values of loan borrowing and its recovery. However, in some

months the forecasts may not agree strongly with the observed values as the series under

study are very volatile and this increases unpredictability in their future values. Therefore,

it was recommended for future researchers to consider non-linear models such as the

ARCH model together with its variants and SVM models in addition to the ARIMA model.