Antony's comments and feedback

[Sayani07]

I have scanned the paper and noticed a few minor issues, listed below. The article is interesting, well structured and well written. Presumably Di and Rob have checked the technical stuff, I have concentrated more on the general and the applications.

• [✅] p4 "Households are distributed geographically and have different demographic properties such as the existence of solar panels, central heating or air conditioning."
• [ ✅] These are more physical properties. Demographic properties would be more about the people living there. Both need to be mentioned.
• [ ✅] p4 "the problem and the solutions we propose are relevant to any temporal data observed more than once per year.” Why not data by years, decades, and centuries? For weather or astronomical data, perhaps.
• [ ✅] p5 What are "prospective probability distributions" ?
• [ ✅] p10 "A classic example is Easter (in the western tradition) whose dates repeat only after 5.7 million years." Fascinating, do you have a reference?
• [❌] §4 What about day of week and week of month? Sometimes there are 6 weeks in a month. Comment: point is the same, for 6 weeks or 5 weeks the pair is a clash
• [✅ ] p18 and in references: Mcgill should be McGill
• [✅ ] p18 Reference for ridge line plots? (Especially as they are new.)
• [✅ ] p18 Displays are not "equipped with benefits". How about "Each has its own strengths and weaknesses and that should be..."
• [ ✅] p20 Fig 4 Why was Q2 unusual?
• [ ✅] Figure 4 caption Plot (c) switches (not switch)
• [ ✅] §5.3 is sensible, but unclear, partially related to numbers of levels in §5.3. Why emphasise transformations, why not just point out that facets with small numbers of cases can be uninformative or potentially misleading?
• [ ❌] p23 Use brackets in the plot caption to Figure 5 as in Figure 4? ie (a) not a. Same for Figure 6. Comment: Everything made consistent by adding borders and labels without brackets
• [ ✅] p25 "that that" should be "that"
• [ ✅] p25 Is the influence of the fall of a wicket not rather that the new batsman needs to play himself in? (So it might matter who is actually facing in the next over and when the wicket falls in the over.)
• [ ✅] p27 the package gs is not easy to find. Should almanac and gs both be referenced?
• [ ✅] p28 §8 check should be checks, recommend should be recommends
• [✅ ] p28 The current version on CRAN is 0.1.3, should that not be used?

[Sayani07]

From Antony

• [ ❌] I have now had a chance to look at the code for the paper. The cricket data are fascinating and I like your first result, although it would be clearer if you used quantile plots rather than lv plots.
  comment: To have range of graphics, keep it to lv plots
• [ ✅] The second result is also good, but why have you only considered dismissals involving being caught and not all dismissals? The results are clearer if you use all