Shortened numbers in Stata

[Jodar Joldoshov]

I am a new user of Stata and don't quite understand how to read numbers when they are shortened in Stata, such as these 1.60e-06, 3.62e-07, 8.91e-07, 2.31e-06 and so on. Could anyone please clarify how I should properly interpret these numbers?

[Leonardo Guizzetti]

These numbers are in what’s called scientific (or standardized) notation. A number takes the format of XeY, which means to take X and multiply by 10 raised to the power of Y, or XeY = X * 10^Y.

[William Lisowski]

Those numbers have been displayed in scientific notation, as discussed for example in

https://en.wikipedia.org/wiki/Scientific_notation

They represent 1.60 * 10^-6 = 0.00000160, 3.62 * 10^-7 = 0.000000362, and so on.

Stata tends to apply a default format %9.0g that allows 8 digits and sign, and in cases where this does not successfully display any significant digits, effectively shifts to scientific notation (which it calls exponential format) %9.2e.

For more on how Stata formats numbers for display, see the discussion in the output of

help format

If the examples you showed were coefficient estimations from a modeling command, you can often produce more attractive results by rescaling, for example, a figure given in dollars to one given in millions of dollars. The coefficient 1.60e-06 would then be displayed as for example 1.60432 or 1.599934 with more significant digits shown.

[daniel klein]

. display %9.8f  1.60e-06
0.00000160

. display %10.9f 3.62e-07
0.000000362

. display %12.9f 8.91e-07
 0.000000891

. display %9.8f  2.31e-06
0.00000231