1 - Considerando uma distribuição padrão Z ~ N(0,1)
1.1 - Pr(-1<Z<2.1)
Pr(-1<Z<2.1) = Pr(-1<Z<0)+Pr(0<Z<2.1) =
Pr(0<Z<1)+Pr(0<Z<2.1) = 0.34134 + 0.48214 = 0.82348
(pnorm(0) - pnorm(-1)) + (pnorm(2.1) -pnorm(0))
## [1] 0.8234803
plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(-1, 2.1)), col=c('#2e669e'),
legend=FALSE)

plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(-1, 0),c(0,2.1)),
col=c('#2e669e','#9e2e82'),
legend=FALSE)

1.2 - Pr(-0,87<Z<1,54)
Pr(-0.87<Z<1.54) = Pr(-0.87<Z<0) + Pr(0<Z<1.54) =
0.30785 + 0.43822 = 0.74607
(pnorm(0) - pnorm(-0.87)) + (pnorm(1.54) -pnorm(0))
## [1] 0.7460696
plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(-0.87, 1.54)), col=c('#ccb8c7'),
legend=FALSE)

plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(-0.87,0),c(0,1.54)),
col=c('tomato4','#438f8b'),
legend=FALSE)

1.3 - Pr(Z=1,54) é igual a zero pois se trata de um ponto
discreto
Pr(Z=1.54) = Pr(0<Z<1.54) - Pr(-1.54<Z<0) = 0.9382198 -
0.9382198 = 0
(pnorm(1.54) - pnorm(1.54))
## [1] 0
plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(0,1.54),c(-1.54,0)),col=c('#263618','#438f8b'),
legend=FALSE)

1.4 - Pr(Z>2,5)
Pr(Z>2,5) = Pr(0>Z>3) - Pr(0>Z>2,5) = 0.4986501 -
0.4937903 = 0.004859767
(pnorm(3) - pnorm(0)) - (pnorm(2.5) - pnorm(0))
## [1] 0.004859767
plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(0,3)),
col=c('#263618'),
legend=FALSE)

plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(0,2.5)),
col=c('#c79540'),
legend=FALSE)

plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(0,3),c(0,2.5)),
col=c('#aba59a','#e6a83e'),
legend=FALSE)

plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(2.5,3)),
col=c('#e6733e'),
legend=FALSE)

2 - Considerando uma distribuição X ~ N(500,300)
2.1 - Pr (X<500)
Padronizando a distribuição:
Pr (Z < (500-500)/300) = 0/300 = Pr (Z<0) = Pr
(-3.291<Z<0)
(pnorm(0) - pnorm(-3.291))
## [1] 0.4995008
plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(-3.291,0)),
col=c('#69625f'),
legend=FALSE)

2.2 - Pr (400<X<710)
Padronizando a distribuição:
Pr ((400-500)/300) < Z < ((710-500)/300) = Pr (-100/300 < Z
< 210/300)
Pr (-0.33 < Z < 0.7) = 0.12930 + 0.2580363 = 0.38734
(pnorm(0) - pnorm(-0.33)) + (pnorm(0.7) - pnorm(0))
## [1] 0.3873364
plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(-0.33,0),c(0,0.7)),
col=c('#7c3c80','#d69f67'),
legend=FALSE)

2.3 - Pr (330<X<450)
Padronizando a distribuição
Pr ((330-500)/300) < Z < ((450-500)/300) = Pr (-0.57 < Z
< -0.17) = 0.2156612 - 0.06749493 = 0.1481662
(pnorm(0) - pnorm(-0.57)) - (pnorm(0) - pnorm(-0.17))
## [1] 0.1481662
plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(-0.57,0),c(-0.17,0)),
col=c('#b327a5','#b32738'),
legend=FALSE)

3 - Considerando uma distribuição Y ~ N(100,20)
3.1 - Pr (50 < Y < 70)
Padronizando a distribuição:
Pr ((50-100)/20) < Z < ((70-100)/20) = Pr (-2.5 < Z <
-1.5) = 0.4937903 - 0.4331928 = 0.06059754
(pnorm(0) - pnorm(-2.5)) - (pnorm(0) - pnorm(-1.5))
## [1] 0.06059754
plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(-2.5,0),c(-1.5,0)),
col=c('tomato4','royalblue'),
legend=FALSE)

3.2 - Pr (105 < Y < 110)
Padronizando a distribuição:
Pr ((105-100)/20 < Z < (110-100)/20) = Pr (0.25 < Z <
0.5) = Pr (0 < Z < 0.5) - Pr (0 < Z < 0.25) = 0.1914625 -
0.09870633
(pnorm(0.5) - pnorm(0)) - (pnorm(0.25) - pnorm(0))
## [1] 0.09275614
plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(0,0.5),c(0,0.25)),
col=c('tomato4','royalblue'),
legend=FALSE)

3.3 - Pr (Y > 75)
Padronizando a distribuição:
Pr (0 > Z (75-50)/20) = Pr (0 > Z > 1,25) = 0.3943502
pnorm(1.25) - pnorm(0)
## [1] 0.3943502
plotDistr(x, dnorm(x, mean=0, sd=1), cdf=FALSE,
regions=list(c(0,1.25)),
col=c('tomato4'),
legend=FALSE)
