Monthly Inflation Dynamics in Somalia Using Seasonal ARIMA Models

In the complex machinery of a national economy, few forces are as immediate or as personally felt as the price of goods. When the cost of bread, fuel, or medicine rises steadily over time, it is called inflation. This phenomenon erodes the purchasing power of money, meaning a family can buy less with the same amount of cash than they could the year before. For governments, predicting these price changes is a matter of survival; accurate forecasts allow them to adjust wages, plan budgets, and protect the most vulnerable citizens from sudden economic shocks. However, in many developing nations, the data needed to make these predictions is often sparse, irregular, or difficult to interpret. In places like Somalia, where markets are fragile and supply chains are easily disrupted by climate or conflict, understanding the rhythm of price changes is not just an academic exercise but a critical tool for stability.

The challenge for economists in such environments is distinguishing between random noise and a predictable pattern. Prices do not move in a straight line; they often rise and fall with the seasons, driven by harvest cycles, religious holidays, or weather patterns. To navigate this, researchers rely on statistical models that treat the economy like a weather system, looking for recurring cycles rather than just random storms. One such approach involves analyzing how prices in one month relate to prices in the same month of the previous year, or how a price spike today might influence prices next month. By mapping these connections, it becomes possible to see the underlying structure of inflation, separating the predictable seasonal swings from the unpredictable chaos of daily life.

This is the precise task undertaken by Abdirashiid Mohamud Mohamed, a researcher at the Somali National University, in a new study focused on the monthly inflation rates of Somalia. The goal was not merely to describe what happened in the past, but to build a reliable tool for looking ahead. Using official data from the Somalia National Bureau of Statistics, covering the period from February 2018 to December 2024, the researcher applied a specific type of statistical framework known as a Seasonal Autoregressive Integrated Moving Average model. In plain terms, this is a method designed to find patterns in data that repeat over a fixed cycle, such as the twelve months of a year. The study focused specifically on the month-to-month rate of price change, rather than the total price level, to capture the speed and direction of inflation more accurately.

The analysis began by checking the stability of the data. The researcher confirmed that the monthly inflation rates were stable enough to be modeled without needing complex mathematical adjustments to force them into a straight line. When the data was examined for patterns, a clear and dominant rhythm emerged: the inflation rate in any given month was strongly linked to the rate in the same month of the previous year. This annual cycle was the most powerful signal in the data, far outweighing the influence of what happened in the immediately preceding month. While there was some short-term persistence, where a price rise this month might nudge prices up next month, the yearly recurrence was the primary driver.

To find the best way to describe this behavior, the researcher tested nine different mathematical structures against the historical data. The winner was a model that combined a simple short-term memory with a strong seasonal memory. This optimal model successfully captured the reality that Somalia's inflation is largely a seasonal phenomenon. The study found that the price changes in a specific month, say March, are heavily determined by what happened in March of the year before. The statistical evidence showed that this seasonal link is strong, positive, and highly significant, meaning that the annual cycle is a reliable feature of the Somali economy, not a random fluke.

With the model validated, the researcher used it to generate forecasts for the years 2025 and 2026. The results painted a picture of an economy that remains relatively stable on average but is marked by distinct, predictable peaks and valleys. The model projects that inflation will tend to rise during the first half of the year, specifically peaking in March, April, and May. These months correspond to periods of seasonal pressure, likely driven by food supply adjustments after the rainy seasons and increased demand during religious observances. Conversely, the model predicts periods of falling prices, or deflation, in January, August, and December. These dips are not necessarily signs of economic weakness but rather part of the natural seasonal ebb and flow of the market.

The accuracy of these predictions was tested against real-world data from 2024, which was held back from the initial analysis to serve as a blind test. The model performed with high precision, with its average error being less than one-third of the typical monthly fluctuation in prices. This level of accuracy suggests that the model has successfully identified the true engine driving inflation in the country. Furthermore, a check on the stability of the model showed that the seasonal pattern has remained consistent over time, with no sudden breaks or shifts that would render the forecasts useless.

The implications of these findings are practical and immediate. For policymakers, the study offers a clear warning: the spring months will likely bring higher prices, and preparations should be made in advance to protect low-income households. The predicted price drops in winter and late summer should not be mistaken for a permanent improvement in the economy but understood as part of the seasonal cycle. While the model provides a solid baseline for planning, the researcher notes that it cannot account for sudden, external shocks like global fuel price spikes or severe droughts, which can disrupt the pattern at any time. Nevertheless, by mapping the predictable rhythms of the economy, this work provides a crucial foundation for more stable and effective economic management in Somalia.