For question 5(e) of the January 2009 S1 paper, what does it mean by 'average and measure of dispersion'. Why do you choose the median/IQR and how does skewness affect this? Thanks for the help!
For question 5(e) of the January 2009 S1 paper, what does it mean by 'average and measure of dispersion'. Why do you choose the median/IQR and how does skewness affect this? Thanks for the help!
So average is just: which one of mean, mode or median should you use? Dispersion is just: which one of IQR or range should you use?
Since the data is positively skewed then the mean is unrepresentative of the overall data as it will be skewed positively whereas the median gives you a nice non-skewed 'middle' of the data, i.e: it's not affected by the extreme positive values present because of the skew.
Same goes for range, the range will be exaggerated by the positive skew that makes it much bigger than it should normally thought of to be. SO using IQR cuts out that high-positive end and is a better measure of how the data is dispersed.
So average is just: which one of mean, mode or median should you use? Dispersion is just: which one of IQR or range should you use?
Since the data is positively skewed then the mean is unrepresentative of the overall data as it will be skewed positively whereas the median gives you a nice non-skewed 'middle' of the data, i.e: it's not affected by the extreme positive values present because of the skew.
Same goes for range, the range will be exaggerated by the positive skew that makes it much bigger than it should normally thought of to be. SO using IQR cuts out that high-positive end and is a better measure of how the data is dispersed.
Thanks so much, that explanation was very helpful!
A continuous distribution has the characteristic that the probability of P(X=anything) = 0 since the probability of picking any single point is 0. When dealing with continuous variables the only times that probabilities are non-zero are when you have intervals, so P(X < 10) or P(5 < X < 7) or w/e. P(X=n) = 0 for all n for a continuous distribution.
Also, for S1 papers what degree of accuracy should you give your answers to?
A continuous distribution has the characteristic that the probability of P(X=anything) = 0 since the probability of picking any single point is 0. When dealing with continuous variables the only times that probabilities are non-zero are when you have intervals, so P(X < 10) or P(5 < X < 7) or w/e. P(X=n) = 0 for all n for a continuous distribution.
Only part(c) is conditional probability and that is because they use the word "given". In general, dependence has nothing to do with conditional probability.
Only part(c) is conditional probability and that is because they use the word "given". In general, dependence has nothing to do with conditional probability.
So when drawing the tree diagram for the second set of branches, trial 2, should the notation be P(B∣C)orP(B)
For question 4(c) of this paper why is it P(L>133∣L>127), I don't understand the L>127 bit. I would've thought it would be equal too, but I know that can't be right as then it would be 0 because it is continuous. Also, why isn't it less than?
For question 4(c) of this paper why is it P(L>133∣L>127), I don't understand the L>127 bit. I would've thought it would be equal too, but I know that can't be right as then it would be 0 because it is continuous. Also, why isn't it less than?
"Given that it is 127 hours since Alice last charged her phone" means that it's been at least 127 hours since Alive charged her phone throughout the journey. So it could be anywhere between 127 hours and infinity hours since she last charged her phone.