Question about volume and number of order #2

dennislx · 2022-01-09T01:14:29Z

Thank you by sharing this valuable dataset to quantitative researchers around the globe.

I have questions about these two lines in your paper:

Volumes are normalized by dividing by the top 10% volume quantile level, but limited to 1

Number of orders are normalized similar to volumes, dividing by the 10% order number
quantile level, but limited to 1.

It seems you are dividing the column with its top 10% quantile. The frequency, however, does not seem to match what you have described in the paper

>>> Counter(data[:, 11])    #this is the 12th column, which is bid volume at level 1
Counter({1.0: 234366,                                                                            
         0.01: 1230,                                                                             
         0.02: 913,                                                                              
         0.03: 847,                                                                              
         ....
}
>>> m = Counter(data[:, 11])
>>> sum(m[k] for k in m if k != 1.0)/len(data)
0.15361085731001334

Other columns with normalization show the same problem to me. I am probably wrong somewhere. Can you please share a separate raw volume data that could help clear my confusions?

Thanks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Question about volume and number of order #2

Question about volume and number of order #2

Question about volume and number of order #2

Question about volume and number of order #2

Comments

Uh oh!