Uživatel:Juandev/Economy of traditional ethnomedicinal markets: Porovnání verzí
Smazaný obsah Přidaný obsah
→Hypotheses, setting hypotheses: update značka: editace z Vizuálního editoru |
→Statistical hypothesis testing: meziuložení značka: editace z Vizuálního editoru |
||
Řádek 1 493:
==== Statistical hypothesis testing ====
* '''null hypothesis''' stay for default hypothesis Ho
** based on preivous studies
**= currentyl accepted value for a parametr
**previous study on that
** everybody things how is it
** its still true unless we will discover testing Ha the Ha is true
* '''alternative''' or '''research
** new hypothesis to be tested
**so it could be mathematical opposite
*results of testing
**reject Ho or fail to reject Ho ~ no way to prove Ha
***if we reject Ho we believe its not true
***if we fail to reject Ho we continue Ho is true
*test statistic = caclulated from the sample data, used to decide
*statisticaly significant = where we do draw the line to make a decision
*level of confidence (=důvěra C) in % = how cofidente are we in our decision
*level of significance (=význam) α = 1 - C %
**for C = 95, C = 0.95, than α = 1 - 0.95 ~ '''α = 0.05'''
* WHY I AM TESTING SOMEONE ELSE FINDINGS THAN?
*testing
**Ho : μ = 4 mm
**Ha : μ ≠ 4 mm
**n = 100 ... one hundred semples to test
**C = 0.99 ...we want 99 % of confidence
***α = 1 - C = 1 - 0.99 = 0.01
*
*
| |||