Teorifrågor matstat Flashcards
Numerical data can be divided into two types: discrete and continuous.
Are these classes overlap, i.e. can a data be discrete and continuous type at the same time?
1 No;
2 Mathematically, no, but for analytical purposes it may be convenient to treat discrete data with a fine grid as continuous
2 Mathematically, no, but for analytical purposes it may be convenient to treat discrete data with a fine grid as continuous
The words above each answer box refer to a data attribute which can be associated with people’s activities.
Identify the type of each attribute; and enter 1 or 2 or 3 or 4 into the appropriate answer box:-
1 for nominal data, 2 for ordinal data, 3 for discrete numerical data, and 4 for continuous numerical data.
If the number of different possible (sensible) numerical values exceeds 20, then (for this question) respond with ‘4’ rather than ‘3’
a Number of goals scored in a football match
b Street you live in
c The time until the end of the class
d Disease incidence: epidemic, common, rare
e Community type : hamlet, village, town, city
a Number of goals scored in a football match
Discrete numerical data
b Street you live in
Nominal Data
c The time until the end of the class Continuous numerical data
d Disease incidence: epidemic, common, rare
Ordinal data
e Community type : hamlet, village, town, city
Ordinal data
A histogram consists of rectangles with the bases spanning the range of the sample divided into disjoint regions: bins. Which measure of rectangles is proportional to the number of observations in the corresponding bins?
It is:
1 height
2 area
3 weight
4 none of these
Area
The Pie chart is considered as a bad presentation tool by serious statisticians.
This is because
1 it raises appetite;
2 statisticians are jealous of the bakers;
3 the humans are bad at comparing non-linear measures: areas of wedges in this case.
The humans are bad at comparing non-linear measures: areas of wedges in this case.
Robustness is the property of being (almost) insenitive to the presence of outliers.
Is the mean of the sample is a robust measure of the location of data?
1 Yes;
2 No;
3 Who cares?
No
Robustness is the property of being (almost) insenitive to the presence of outliers.
Is the IQR of the sample is a robust measure of the data variability?
1 Yes;
2 No;
3 Who cares?
No
Which measure of variability is preferable to use for highly skewed data?
It is better to use
1 Standard deviation;
2 Mode;
3 IQR
IQR
Is it possible to speak about probability that the 1st of January 2027 will be snowy?
a The answer is
1 yes, why not?;
2 no, because it is not a repeatable experiment;
3 I wish it does!
b How should one reformulate the question to make probabilistic sense to it?
1 It snows in Gothenburg on the 1st of January;
2 It always snows on the 1st of January;
3 It rains on the 1st of January 2017.
a no, because it is not a repeatable experiment;
b It snows in Gothenburg on the 1st of January;
What does it mean formally
a that an event E happens?
1 That the observed elementary outcome is an element of E;
2 That the observed elementary outcome is not an element of E;
3 That the observed elementary outcome is an element of the sample space
b that an event E does not happen?
1 That the observed elementary outcome is an element of E;
2 That the observed elementary outcome is not an element of E;
3 That the observed elementary outcome is an element of the sample space
a That the observed elementary outcome is an element of E;
b That the observed elementary outcome is not an element of E;
The rules of Probability concerning the unions and intersections of sets (events)
is similar to those of
1 length 2 area 3 volume 4 electrical charge 5 all of these
All of these
Let A and B be two events happening with positive probability.
a How do P(A) and P(A|B) compare?
1 No relation 2 P(A) > P(A|B) 3 P(A) < P(A|B) 4 P(A) = P(A|B)
What does P(A|B) > P(A) imply?
1 Nothing 2 P(B|A) 3 P(B|A)>P(B) 4 P(AB)=0
a No relation b P(B|A)>P(B)
Let A,B,C be events such that P(ABC) > 0.
What is P(ABC) always equal to?
1 0 2 1 3 P(A) P(B) P(C) 4 P(A|B) P(B|C) P(C|A) 5 P(A) P(B|A) P(C|AB)
5 P(A) P(B|A) P(C|AB)
Let B,C and D be events of positive probabilities.
For the Full Probability formula
P(A)=P(A|B)P(B)+P(A|C)P(C)+P(A|D)P(D)
to be true, it is necessary and sufficient that B,C,D are…
1 disjoint 2 covering the whole sample space 3 of equal probabilities 4 forming a partition of the sample space 5 independent
4 forming a partition of the sample space
Let positive probability events A and B be such that P(A|B)=P(A|Bc).
Which of the following are then true?
- P(A) = P(B)
- P(A) = P(A|B)
- A and B are independent
- P(B) = 1/2
1 Only 1 2 Only 2 3 Only 3 4 Only 4 5 1 and 2 6 1 and 3 7 1 and 4 8 2 and 3 9 2 and 4 10 3 and 4 11 All 12 None
- 2 and 3
What is a random variable?
In mathematical terms, a discrete random variable is
1 a variable which value is unknown
2 a variable which value is random
3 a function on the sample space
4 it is not a mathematical notion
- a function on the sample space