Hello Yoganageswarry,

I'm a bit unclear as to your specific goal, so this response may not be as focused as you might prefer.

If your aim is to compare the trajectory of the two time series (for crime "A" and crime "B" types), you could:

1. Regress frequency on year for crime "A" and likewise for crime "B", and test whether the regression coefficients were equal for the two models (probably have to test linear, quadratic, cubic, etc. models).

2. Run a chi-square test of independence for crime type and year, compiling the frequencies as the cell observations for the 2 x 18 table.

3. Use the Kolmogorov-Smirnov goodness of fit test to compare the distributions of cases over time. You'd be constructing two cumulative relative frequency distributions (e.g., time 1 = year 2000 cases only; time 2 = year 2000 + year 2001 cases...).

If you have some other aim in mind, perhaps you could elaborate your query.

Good luck with your work.

Professor Morse's advice is certainly correct. However Option 1 is not necessarily a simple nor polynomial fit. I would start with some scatter plots to investigate possibilities of other models. As Morse suggested a research question would be of assistance in answering your question. Good luck 🤞🤞, David Booth

You can check the location of spread of the two data sets are same or not(Non parametric tests)or mean or variance (using non parametric tests).Thank you