search

bobyellow / BiFlowLISA

Bivariate local Moran's I for spatial flow data
MIT License
3 stars 0 forks source link

Questions about data #2

Open MHKIM7 opened 4 months ago

MHKIM7 commented 4 months ago

Dear Prof. Tao,

Hello,

I am grateful for the BiFlow LISA code and the paper you published, as they have been extremely helpful.

I am writing to ask a question that came up during my analysis.

I have conducted a BiFlow LISA analysis and seek a review of the results for potential errors. I am interested in identifying any mistakes or gaps in my understanding of the theory. An email has been sent to you concerning this issue, and I would be grateful for your assistance if you are available.

Thank you for your attention.

20240724 Kim

bobyellow commented 3 months ago

Hi Kim, Thank you for your interest in my paper. The data on Github was meant for practicing the codes, not necessarily exactly the same as in my paper (probably the same data but with further processing). Do you have to use the same data in my paper for testing? If you have issues with running the codes with your data, let me know. best, Ran

On Tue, Jul 23, 2024 at 9:20 PM MHKIM7 @.***> wrote:

Dear Prof. Tao,

Hello,

I am grateful for the BiFlow LISA code and the paper you published, as they have been extremely helpful.

I am writing to ask a question that came up during my analysis.

I am curious to know if the flow data for taxis and ride-hailing used in your BiFlow LISA paper is the same as the data uploaded on GitHub.

  1. The paper mentions 2,570 non-NULL flows for both modes. However, the taxi data on GitHub has 2,520 flows, and the ride-hailing data has 2,557 flows.

    • The data on GitHub seems to have already been cleaned of NULL values. For instance, in the taxi data, there is no record of a trip from origin 1 to destination 2, indicating that any NULL entries for this route have likely been eliminated.

  2. There is a discrepancy between the 2,570 flows mentioned in the paper and the 2,520 taxi flows and 2,557 ride-hailing flows on GitHub. Confirmation is needed to determine if these datasets are the same or if there has been an omission.

If you have time, I have some questions about my analysis results that I would like to ask. It would be incredibly helpful if you could review them via email.

Thank you for your attention.

20240724 Kim

— Reply to this email directly, view it on GitHub https://github.com/bobyellow/BiFlowLISA/issues/2, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABUMKQ35OSSANCFXTSEGQUTZN36PDAVCNFSM6AAAAABLLPDS52VHI2DSMVQWIX3LMV43ASLTON2WKOZSGQZDMMZZGE3DONI . You are receiving this because you are subscribed to this thread.Message ID: @.***>