Bioequivalence and Bioavailability Forum

Rounding [Software]

posted by d_labes  – Berlin, Germany, 2013-01-05 20:25 (4914 d 05:58 ago) – Posting: # 9787
Views: 15,073

Dear Helmut!

❝ ... In the meantime I finished my balanced cross-over sim’s. n 12–60, CV 6–100%. With my grid I have 306·10⁶ sim’s each. With rounding I got 24 (7.84%) empiric alphas significantly >0.05; without 11 (3.59%).

How do you interpret the un-rounded results?
Empirical evidence of an Alpha inflation of TOST ?
Hopefully not.

❝ Sometimes R drives me nuts.

Here I have another one:
round(c(1.2500499999999999, 1.250049999999999999), 4)
[1] 1.2500 1.2501

More trouble spots can be found in "The R inferno".
Should be a must-read for all R-users.

Complete thread:

• Rounding Helmut 2013-01-01 16:51 
  • Rounding d_labes 2013-01-02 12:51
    • Rounding Helmut 2013-01-04 17:35
      • Roundingd_labes 2013-01-05 19:25
        • Rounding Helmut 2013-01-05 20:13
          • Sim’s are sim’s are sim’s d_labes 2013-01-05 20:57
        • Another vicious circle Helmut 2013-01-06 02:21
          • Where all these numbers came from? d_labes 2013-01-07 15:52
            • Flaw in the GL? Helmut 2013-01-07 17:29
              • Flaw in the GL? d_labes 2013-01-08 11:44
                • What a mess! Helmut 2013-01-08 19:08
                  • What a mess! d_labes 2013-01-09 10:33
                    • What a mess! Helmut 2013-01-09 15:18
  • Rounding ElMaestro 2013-01-02 16:12
    • Abandon rounding Helmut 2013-01-02 17:04
  • Rounding yjlee168 2013-01-05 23:15
    • Rounding Helmut 2013-01-06 03:29
    • Rounding Helmut 2013-01-06 15:34
      • Rounding yjlee168 2013-01-06 23:38