Bioequivalence and Bioavailability Forum

Mathematical wizard [Bioanalytics]

posted by Helmut  – Vienna, Austria, 2009-01-20 18:03 (5572 d 04:27 ago) – Posting: # 3076
Views: 25,143

Hi Martin,

I really loved one of the examples from your last reference:
  STDEV(1,2,3) = 1
  STDEV(90000001,90000002,90000003) = 0
which looks even more funny if you run this set (in M$-Excel 5.0 and 2000 – the only versions I have handy):
  STDEV(0.90000001,0.90000002,0.90000003) = 0
  STDEV(0.9000001,0.9000002,0.9000003) = 0.0000001
  STDEV(0.900001,0.900002,0.900003) = 0.000001
  STDEV(0.90001,0.90002,0.90003) = 0.00001
  STDEV(0.9001,0.9002,0.9003) = 0.0001
  STDEV(0.901,0.902,0.903) = 0.001
  STDEV(0.91,0.92,0.93) = 0.01
  STDEV(9.1,9.2,9.3) = 0.1
  STDEV(91,92,93) = 1
  STDEV(901,902,903) = 1
  STDEV(9001,9002,9003) = 1
  STDEV(90001,90002,90003) = 1
  STDEV(900001,900002,900003) = 1
  STDEV(9000001,9000002,9000003) = 1
  STDEV(90000001,90000002,90000003) = 0
  STDEV(9000000001,9000000002,9000000003) = 0
  STDEV(90000000001,90000000002,90000000003) = 0
  STDEV(900000000001,900000000002,900000000003) = 0
  STDEV(9000000000001,9000000000002,9000000000003) = 131072
  STDEV(90000000000001,90000000000002,90000000000003) = 0
  STDEV(900000000000001,900000000000002,900000000000003) = 0

R 2.8.1
  sd(c(90000001,90000002,90000003)) = 1
STATISTICA 5.1H
  St.Dev.(90000001,90000002,90000003) = 1
NCSS 2001
  Standard Deviation (90000001,90000002,90000003) = 1
Maxima 5.15.0*
  std1([90000001,90000002,90000003]); = 1
GNU Octave 3.0.3
  std([90000001,90000002,90000003]) = 1
WinNonlin 5.2.1
  STDEV(90000001;90000002;90000003) = 0

As you said, it’s advisable to use statistical software for statistical calculations (surprisingly PK software failed this simple test as well).
Since this thread is in the Bioanalytics category: as far as I know Thermo’s Xcalibur software has only the options to store data in its own proprietary format or to export xls-files.
But help is on the way. All (?) statistical software are able to import xls-files.

---

• When it comes to numerical precision, it’s nice to have a clever algorithm.
  Maxima gives a standard deviation of 1 even for a set of {9·10^1,000+1, 9·10^1,000+2, 9·10^1,000+3}…
  If you don't believe me, run this code in Maxima:
  load(descriptive)$
std1(([9,9,9])*10^1000+([1,2,3]));

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна! 
Helmut Schütz


The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Complete thread:

• Excel sheet usage in bioanalytical calculations chbs 2009-01-19 10:35 [Bioanalytics] 
  • do not use excel martin 2009-01-19 11:52
    • do not use excel ? Ohlbe 2009-01-19 23:54
      • Never martin 2009-01-20 14:52
        • Mathematical wizardHelmut 2009-01-20 17:03
          • Do not trust in computers d_labes 2009-01-21 11:00
            • Do not trust in computers Helmut 2009-01-21 13:00
              • Do not trust in computers Ohlbe 2009-01-21 14:03
                • Do not trust in Excelence d_labes 2009-01-21 15:42
                • Do not trust in computers Helmut 2009-01-22 02:49
                  • Do not trust in computers Ohlbe 2009-01-22 10:33
              • Even my calculator... Ohlbe 2009-01-22 22:06
                • Even my calculator... Helmut 2009-01-22 23:26
                  • Even my calculator... ElMaestro 2009-01-23 12:12
                    • Even my calculator... Ohlbe 2009-01-24 19:44
          • data exchange format martin 2009-01-21 12:16
            • Raw data (paper vs. electronic) Helmut 2009-01-21 12:31
          • Mathematical wizard - use WinNonlin Tools menu instead SDavis 2009-01-22 11:56
      • Why not use R? Helmut 2009-01-21 01:02
        • Why not use R? ElMaestro 2009-01-22 19:32
        • Why not use R? ElMaestro 2009-01-23 22:45
          • Why not use R? Aceto81 2009-01-26 16:15
            • Why not use R? d_labes 2009-01-26 16:35