Question

I’m going crasy with the results from MySql regarding standard functions:
– AVG() the average
– STD() the standard deviation

Check the following results from my table ‘Auction’:

mysql> SELECT avg(buyout) avg FROM auction where buyout <> 0 and item =72988;
+-------------+
| avg         |
+-------------+
| 234337.3622 |
+-------------+

That result looks correct, no issue.
But when I run std:

mysql> SELECT std(buyout) std FROM auction where buyout <> 0 and item =72988;
+-------------+
| std         |
+-------------+
| 574373.6098 |
+-------------+

! The SDT is greater than AVG (SDT > AVG), and that’s… impossible because my AVG>0.

Where am I wrong here … ?
thx in advance !

Answer 1

There is no mathematical constraint saying that if mean is positive it has to be smaller than the standard deviation.

I read the extract of your data in R

data <- read.table("extract_72988.csv", h=1, sep="\t")

And confirmed that

> mean(data$BUYOUT)
[1] 234337.4
> sd(data$BUYOUT)
[1] 574421.3

Further analysis of your data shows that it is far from being normally distributed

Here is an histogram of your data:

And here is the histogram of log-transformed data

And finally a normal Q-Q plot