Size: 610
Comment:
|
Size: 2325
Comment:
|
Deletions are marked like this. | Additions are marked like this. |
Line 5: | Line 5: |
Add your row to the table below, following the examples already there. Column 1 = name. Columns 2 + 3 = p-value for Z-test for hex vs. integer ids with 3 and 10 measurements. Columns 4 + 5 = same with T-test. Please write all p-values as percentages with exactly one digit after the dot. | Add your row to the table below, following the examples already there. Column 1 = name. Columns 2 + 3 + 4 + 5 + 6 = values for n = 3, namely: difference M of means, estimated standard deviation S' = S / sqrt(n), value of test statistic x = M / S', and the p-values for the Z-test and the T-test, respectively. Columns 7 + 8 + 9 + 10 + 11 = same values for n = 10. Please write all values with exactly three digits after the dot. |
Line 7: | Line 7: |
||'''Name''' ||'''Z-test, n = 3'''||'''Z-test, n = 10''' ||'''T-test, n = 3''' ||'''T-Test, n = 10''' || ||Jacob C. ||1.0% || 0.1% || 20.0% || 2.0%|| |
Recall: experiment to measure query times for tables with hex vs. integer ids for person names when joining ''acted_in'' and ''has_won'' on their respective first column. Note: in Wolfram Alpha you can type something like ''P(X >= x) for the standard normal distribution'' or ''P(X >= x) for the t-distribution with ... degrees of freedom'', awesome! ||<style="text-align:center" |2>'''Name''' ||||||||||<style="text-align:center">'''n=3''' ||||||||||<style="text-align:center">'''n=10''' || ||'''M''' ||'''S' = S/sqrt(n)''' ||'''x = M / S'''' ||'''Pr(Z >= x)''' ||'''Pr(T >= x)''' ||'''M''' ||'''S' = S/sqrt(n)''' ||'''x = M / S'''' ||'''Pr(Z >= x)''' ||'''Pr(T >= x)''' || ||Florian B ||0.077 ||0.007 ||10.286 ||0.000% ||0.002% ||0.072 ||0.009 ||7.426 ||0.000% ||0.000% || ||Nikolaus M ||0.253 ||0.0078 ||32.572 ||0.000% ||0.000% ||0.247 ||0.0071 ||34.621 ||0.000% ||0.000% || ||Mallikarjun ||0.507 ||0.0075 ||67.05 ||0.000% ||0.000% ||0.509 ||0.0076 ||66.32 ||0.000% ||0.000% || ||Severin ||0.3494 ||0.0110 ||31.772 ||7.864e-220% ||5.849e-04% ||0.3496 ||0.0071 ||49.288 ||0.000% ||2.2e-16% || ||Chris S ||0.249 ||0.012 ||19.595 ||0.000% ||0.000% ||0.263 ||0.025 ||10.313 ||0.000% ||0.000% || ||Tobias F ||0,1452 ||0,001 ||2,5577 ||0.0054% ||4.000% ||0,1328 ||0,0026 ||0,8132 ||?% ||?% || ||Robin S ||0.147 ||0.005 ||32.439 ||3.871e-229% ||2.854e-6% ||0.176 ||0.012 ||14.414 ||2.113e-45% ||2.503e-10% || ||Andreas H ||0.009 ||0.002 ||4.287 ||0.000 ||0.012 ||0.010 ||0.001 ||9.071 ||0.000 ||0.000 || ||Matthias H ||0.373 ||0.00002 ||134.516 ||0.000% ||0.000% ||0.371 ||0.003 ||108.402 ||0.000% ||0.000% || ||Marc P ||0.3273 ||0.0083 ||32.18 ||0.000% ||0.000% ||0.3243 ||0.0033 ||43.73 ||0.000% ||0.000% || |
Results for Exercise Sheet 13 (Statistical significance)
Please read the instructions below, before adding something to the table!
Add your row to the table below, following the examples already there. Column 1 = name. Columns 2 + 3 + 4 + 5 + 6 = values for n = 3, namely: difference M of means, estimated standard deviation S' = S / sqrt(n), value of test statistic x = M / S', and the p-values for the Z-test and the T-test, respectively. Columns 7 + 8 + 9 + 10 + 11 = same values for n = 10. Please write all values with exactly three digits after the dot.
Recall: experiment to measure query times for tables with hex vs. integer ids for person names when joining acted_in and has_won on their respective first column.
Note: in Wolfram Alpha you can type something like P(X >= x) for the standard normal distribution or P(X >= x) for the t-distribution with ... degrees of freedom, awesome!
Name |
n=3 |
n=10 |
||||||||
M |
S' = S/sqrt(n) |
x = M / S' |
Pr(Z >= x) |
Pr(T >= x) |
M |
S' = S/sqrt(n) |
x = M / S' |
Pr(Z >= x) |
Pr(T >= x) |
|
Florian B |
0.077 |
0.007 |
10.286 |
0.000% |
0.002% |
0.072 |
0.009 |
7.426 |
0.000% |
0.000% |
Nikolaus M |
0.253 |
0.0078 |
32.572 |
0.000% |
0.000% |
0.247 |
0.0071 |
34.621 |
0.000% |
0.000% |
Mallikarjun |
0.507 |
0.0075 |
67.05 |
0.000% |
0.000% |
0.509 |
0.0076 |
66.32 |
0.000% |
0.000% |
Severin |
0.3494 |
0.0110 |
31.772 |
7.864e-220% |
5.849e-04% |
0.3496 |
0.0071 |
49.288 |
0.000% |
2.2e-16% |
Chris S |
0.249 |
0.012 |
19.595 |
0.000% |
0.000% |
0.263 |
0.025 |
10.313 |
0.000% |
0.000% |
Tobias F |
0,1452 |
0,001 |
2,5577 |
0.0054% |
4.000% |
0,1328 |
0,0026 |
0,8132 |
?% |
?% |
Robin S |
0.147 |
0.005 |
32.439 |
3.871e-229% |
2.854e-6% |
0.176 |
0.012 |
14.414 |
2.113e-45% |
2.503e-10% |
Andreas H |
0.009 |
0.002 |
4.287 |
0.000 |
0.012 |
0.010 |
0.001 |
9.071 |
0.000 |
0.000 |
Matthias H |
0.373 |
0.00002 |
134.516 |
0.000% |
0.000% |
0.371 |
0.003 |
108.402 |
0.000% |
0.000% |
Marc P |
0.3273 |
0.0083 |
32.18 |
0.000% |
0.000% |
0.3243 |
0.0033 |
43.73 |
0.000% |
0.000% |