Biology Reference
In-Depth Information
(f)
> quit(save = "yes")
1.5
Consider the
gaussian.test
data set included in bnlearn.
(a) Print the column names.
(b) Print the range and the quartiles of each variable.
(c) Print all the observations for which
A
falls in the interval
[
3
,
4
]
and
B
in
.
(d) Sample 50 rows without replacement.
(e) Draw a bootstrap sample (e.g., sample
5000
observations with replacement)
and compute the mean of each variable.
(f) Standardize each variable.
(
−
∞
,−
5
]
∪
[
10
,
∞
)
(a)
> colnames(gaussian.test)
> names(gaussian.test)
(b)
> for (var in names(gaussian.test))
+ print(range(gaussian.test[, var]))
> for (var in names(gaussian.test))
+ print(quantile(gaussian.test[, var],
+ probs = (1:3)/4))
(c)
> condA = (gaussian.test[, "A"] >= 3) &
+ (gaussian.test[, "A"] <= 4)
> condB = (gaussian.test[, "B"] <= -4) |
+ (gaussian.test[, "B"] >= 4)
> gaussian.test[condA & condB, ]
(d)
> gaussian.test[sample(50, replace = FALSE), ]
(e)
> colMeans(gaussian.test[
+ sample(5000, replace = TRUE), ])
(f)
> scale(gaussian.test)
1.6
Generate a data frame with
100
observations for the following variables:
(a) A categorical variable with two levels,
low
and
high
. The first
50
obser-
vations should be set to
low
, the others to
high
.
(b) A categorical variable with two levels,
good
and
bad
, nested within the
first variable, i.e., the first
25
observations should be set to
good
, the second
25
to
bad
,andsoon.
(c) A continuous, numerical variable following a Gaussian distribution with
mean
2
and variance
4
when the first variable is equal to
low
and with
mean
4
and variance
1
if the first variable is equal to
high
.
In addition, compute the standard deviation of the last variable for each con-
figuration of the first two variables.
The variables can be generated as follows:
(a)
> A = factor(c(rep("low", 50), rep("high", 50)),
+ levels = c("low", "high"))
(b)
> nesting = c(rep("good", 25), rep("bad", 25))
> B = factor(rep(nesting, 2),
+
levels = c("good", "bad"))
Search WWH ::
Custom Search