Study your flashcards anywhere!
Download the official Cram app for free >
 Shuffle
Toggle OnToggle Off
 Alphabetize
Toggle OnToggle Off
 Front First
Toggle OnToggle Off
 Both Sides
Toggle OnToggle Off
 Read
Toggle OnToggle Off
How to study your flashcards.
Right/Left arrow keys: Navigate between flashcards.right arrow keyleft arrow key
Up/Down arrow keys: Flip the card between the front and back.down keyup key
H key: Show hint (3rd side).h key
A key: Read text to speech.a key
50 Cards in this Set
 Front
 Back
formula for SD

square root [ average {entries squared} ]


r

average of { x in SU X y in SU}


rms error

square root of (1r squared)


formula for r graph

y = r (SDy/SDx) x + b


addition rule?

add probabilities when mutually exclusive to find the chance that at least one thing happens


multiplication rule?

multiply probabilities when independent to find out the chance that BOTH happen


weird formula to find out exact chance?

binomial formula


law of averages

the larger number of tosses, likely to get closer to expected value but less like to be exactly expected value (error large in absolute terms, but small relative to number of tosses)


EV

number of draws x average of the box


SE (normal  when drawing at random with replacement from a box with numbered tickets)

square root of draws X (SD of box)


what happens if you want to find the probability for a box whose contents do not follow the normal curve

doesn't matter, probability will follow nomral curve even if contents don't


problems with sampling

sampling bias, nonresponder bias


EV in sampling

population percentage


SEpercentage

SE = (SE for #/ sixe of sample) x 100%


how sample size impacts accuracy

JUST ABSOLUTE SIZE


bootstrap method?

use observed results (fracts) as true fracts


EVaverage

average of box


SEaverage

SE for sum / # draws


SEsum

square root (draws) x (SD box)


SE average

Se sum/ draws


SE count

Se for sum (10 box)


SE%

Se count/ # draws x 100%


z

observedexpected (z says how many SES an observed value is from its expected valkue


P is less that 5%

statistically significant


P is less that 1%

highly significant


SD +

square root (number of measurements/(number of measurements  1)) x SD


degrees of freedom (2 sample z test)

# measurements  1


SE difference (if independent)

square root (SE1 squared + SE2 squared)


zdifference

observed difference  expected difference / SE for diference


chi squared

sum of ((observed frequencyexpected frequency)squared/ expected frequency)


bootstrap method?

use observed results (fracts) as true fracts


EVaverage

average of box


SEaverage

SE for sum / # draws


SEsum

square root (draws) x (SD box)


SE average

Se sum/ draws


SE count

Se for sum (10 box)


SE%

Se count/ # draws x 100%


z

observedexpected (z says how many SES an observed value is from its expected valkue


P is less that 5%

statistically significant


P is less that 1%

highly significant


SD +

square root (number of measurements/(number of measurements  1)) x SD


degrees of freedom (2 sample z test)

# measurements  1


SE difference (if independent)

square root (SE1 squared + SE2 squared)


zdifference

observed difference  expected difference / SE for diference


chi squared

sum of ((observed frequencyexpected frequency)squared/ expected frequency)


degrees free (CHI squared test)

# of terms in chisquared  1


Chi squared test to find out how likely observed data is when there is box of tickets w/ contents given draws made @ random w/replacement

chi squared formula, degrees free = terms in chi squared  1


chi squared to show how data was fudged

chi squared/ degrees of freedom can be added up


chi squared to test independence

1  add up all and make percentages (on whole) then calculate expected
2  compute x2 3 degrees of freedom (m1) x (n1) 

a note on tests of singificance

arbitrary (lead to cheating)
