What's the probability event will occur at a specific hour of day? I have an event, let's say page views, that can occur at any time of day. I have a table showing the page views distribution across 24 hours, with time being presented in 1HR increments. With a Chi Square, I can say whether the distribution is evenly distributed or not. But what statistic would I need in order to say, there's a X probability that a person will view the page between 2PM-3PM?
 A: From what I understand, you have data with counts of page views at different times of a day. You are interested in calculating the probability of a page view at a particular hour of the day. Unless I am missing something, you can use this data to calculate the empirical distribution of page views and use it for calculating the probabilities. Just divide the counts at a particular hour by the total of the counts for the 24h, and this is your estimate of the probability of page view at this hour. For this calculation to hold, your data either needs to record all the page views or have a random sample of the views, otherwise, you are risking that the data is biased and it does not give you a clear picture of the counts. Another thing to consider is how representative is the data you have. For example, maybe you collected the data on the weekend and it is not representative for page views during the working days, or maybe you collected it in the summer, and the user activity during winter differs. You would like the data to be representative of the period for which you are making the prediction. You would probably also like to have data collected during different periods (parts of the day, weekdays, months, seasons, etc) so that you could use a statistical model that corrects the seasonalities when calculating the probabilities. But the simplest approach is just using the empirical probabilities.
