David Sütő

Hotspot analysis helps in identifying clustered data points in a map Hence, in performing such hot spot analysis, you could average the results based on cumulative monthly averaging , or using the seasonal averaging. This means that a high z-score value and a low p-value for each month within the three years indicate a significant hotspot. Also, a low negative z-score value and a small p-value indicates a significant cold spot for the same period. Therefore, the higher (or lower) the z-score value, the more intense the clustering. A z-score value near 0 would then connote no spatial clustering.

Calculate the mean of rasters of the required months and then do hotspot analysis over it.

For eg - you want hotspots from September to November. And you have separate rasters of September, October, and November over which you have done hotspot analysis separately. So you can calculate the mean of rasters of September, October and November using cell statistics, and then you can do hotspot analysis over the mean raster of 3 months.

I hope it helps.

The aggregation method to use to create weighted features for analysis from incident point data.

https://desktop.arcgis.com/en/arcmap/10.3/tools/spatial-statistics-toolbox/optimized-hot-spot-analysis.htm

Time series analysis will be okay for the collected data to forecast for the future occurrence and to identify the trend of events in accordance to the month of occurrence and to be able to make appropriate plan towards eradication of such occurrence if it is not favourable