High standard errors in qpDstat results

[Han-Shi-China]

Hi, I noticed that sometimes when I ran qpDstat(version: 662) with "printsd: YES", the output standard error (in column seven) can reach up to one, and the value of D/Z did not equal to the output standard_error. As the following results showed, when the output standard error is 1, the absolute of Z score still reach above 3.

result: Mbuti Afanasievo A B -0.0050 1.000000 -4.005 53480 54018 1099497
result: Mbuti AfontovaGora3 A B -0.0007 1.000000 -0.277 12366 12384 264327
result: Mbuti Aigyrzhal_BA A B -0.0061 1.000000 -3.680 39145 39627 811035
result: Mbuti Altaian.DG A B -0.0035 1.000000 -2.018 54744 55131 1090835
result: Mbuti Ami.DG A B -0.0058 1.000000 -3.473 55438 56082 1092691
result: Mbuti Anatolia_EBA A B -0.0025 1.000000 -1.588 46813 47045 989892
result: Mbuti Anatolia_N A B -0.0046 1.000000 -3.622 53485 53981 1109335
result: Mbuti Andronovo A B -0.0038 1.000000 -1.746 24760 24949 525918

So I am wondering, is this the real calculated standard error, or is this a sign of some calculating problem? Also, are these D results trust-worthy to infer population relationships? In which kind of scenario will this (high standard error) be more likely to happen?
Looking forward to your reply. If there is any thing that I misundertand, I will be really grateful to know. Thanks a lot!!

Han

[bumblenick]

This looks like a flat out bug; Never seen this before but I will look into it. The Z-scores should be OK. Nick

…

On Mon, Jul 15, 2024 at 9:11 AM Han Shi ***@***.***> wrote: Hi, I noticed that sometimes when I ran qpDstat(version: 662) with "printsd: YES", the output standard error (in column seven) can reach up to one, and the value of D/Z did not equal to the output standard_error. As the following results showed, when the output standard error is 1, the absolute of Z score still reach above 3. result: Mbuti Afanasievo A B -0.0050 1.000000 -4.005 53480 54018 1099497 result: Mbuti AfontovaGora3 A B -0.0007 1.000000 -0.277 12366 12384 264327 result: Mbuti Aigyrzhal_BA A B -0.0061 1.000000 -3.680 39145 39627 811035 result: Mbuti Altaian.DG A B -0.0035 1.000000 -2.018 54744 55131 1090835 result: Mbuti Ami.DG A B -0.0058 1.000000 -3.473 55438 56082 1092691 result: Mbuti Anatolia_EBA A B -0.0025 1.000000 -1.588 46813 47045 989892 result: Mbuti Anatolia_N A B -0.0046 1.000000 -3.622 53485 53981 1109335 result: Mbuti Andronovo A B -0.0038 1.000000 -1.746 24760 24949 525918 So I am wondering, is this the real calculated standard error, or is this a sign of some calculating problem? Also, are these D results trust-worthy to infer population relationships? In which kind of scenario will this (high standard error) be more likely to happen? Looking forward to your reply. If there is any thing that I misundertand, I will be really grateful to know. Thanks a lot!! Han — Reply to this email directly, view it on GitHub <#105>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEE77BYRSX5ROTAP55V6SDLZMPC6VAVCNFSM6AAAAABK4QXJMCVHI2DSMVQWIX3LMV43ASLTON2WKOZSGQYDQNZRGA4TQMY> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[Han-Shi-China]

Great, thanks for checking on this! Please let me know when you figure this out, thanks :)