Time Series-Temperature differences

#1
Hello, I have the maximum temperature and the minimum temperature for all days of 2019 in a specific coordinate. Also, I have the differences between them. What I would like to proove (or not) is that the temperature difference in August and September is significantly higher than the rest of the year. I had the idea to take the mean and standard deviation of each month and use a test with the hypothesis that they are equal. However, I do not know if it is entirely correct or if there is a better approach to this problem. Any suggestion?
 

Miner

TS Contributor
#2
There are several approaches you could try:
  • ANOM on the monthly means & StdDev. The ANOM null hypothesis is that no mean is different from the group mean. This is different from the ANOVA null hypothesis, which is that no single mean is different from another mean.
  • Individuals control chart on the daily deltas. This would detect a shift in the daily delta from the baseline.