Styjun

Create an outline of font

Sort a list of pairs representing an acyclic, partial automorphism

Windows 10: How to Lock (not sleep) laptop on lid close?

Is this wall load bearing? Blueprints and photos attached

Typeface like Times New Roman but with "tied" percent sign

Can withdrawing asylum be illegal?

Who or what is the being for whom Being is a question for Heidegger?

How to politely respond to generic emails requesting a PhD/job in my lab? Without wasting too much time

What information about me do stores get via my credit card?

How to delete random line from file using Unix command?

How can I protect witches in combat who wear limited clothing?

The variadic template constructor of my class cannot modify my class members, why is that so?

does high air pressure throw off wheel balance?

First use of “packing” as in carrying a gun

Why does the Event Horizon Telescope (EHT) not include telescopes from Africa, Asia or Australia?

How should I replace vector<uint8_t>::const_iterator in an API?

Semisimplicity of the category of coherent sheaves?

Is it ok to offer lower paid work as a trial period before negotiating for a full-time job?

How to stretch delimiters to envolve matrices inside of a kbordermatrix?

How are presidential pardons supposed to be used?

Working through the single responsibility principle (SRP) in Python when calls are expensive

"... to apply for a visa" or "... and applied for a visa"?

Change bounding box of math glyphs in LuaTeX

Scientific Reports - Significant Figures

Efficient ways to summarize array in R depending on other data

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty height:90px;width:728px;box-sizing:border-box;

0

I want to summarize outcomes from a 3-dimensional array contingent on information from two other datasets. Say, the number of individuals i who after flipping k coins in year t (array) have at least 1 head, with results organized by the individual's sex (vector) and the coin they used for each flip, dime or quarter (matrix). What is the best way to achieve this?

Below are two approaches I tried. Although they seem to work, they take too long to scale...

Let A be the array storing the coin flips, X the vector storing people's sex, and Y the matrix storing the coins used:

A <- array(sample(c("H","T"), size=n.i*n.t*n.k, replace=T), dim=c(n.i, n.t, n.k))
X <- as.logical(rbinom(n.i, 1, 0.49))
Y <- matrix(as.logical(rbinom(n.i*n.p, 1, 0.3)), nrow=n.i, ncol=n.k)

In my case n.i <- 10^5 n.t <- 10^2 n.k <-10

Approach 1 — Vectorized approach:

result <- matrix(0, nrow=n.t, ncol=4)
count <- matrix(0, nrow=n.i, ncol=n.t) 
heads <- A=="H"
for (x in 0:1) # male or female
 for (y in 0:1) # dime or quarter
 count <- 0
 for (k in 1:n.k) 
 count <- count + 1*(X==x & Y[,k]==y & heads[,,k])
 
 result[,1+x+2*(y-1)] <- colSums(count>0) 
 

Approach 2 — Expand X and Y to similar dimensions as A, fixing values along the k and t axis, eg: X <- array(X, dim=c(n.i, n.t, n.k)). Then use apply():

for (x in 0:1) 
 for (y in 0:1) 
 result[,1+x+2*(y-1)] <- apply(apply(X==x & Y==y & heads, 3, sum)>0, 2, sum) 
 
 

Anyone any better solutions?

data-management

share|improve this question

edited Mar 25 at 18:31

Reinier Meester

asked Mar 22 at 6:25

Reinier MeesterReinier Meester

12

add a comment | 

0

I want to summarize outcomes from a 3-dimensional array contingent on information from two other datasets. Say, the number of individuals i who after flipping k coins in year t (array) have at least 1 head, with results organized by the individual's sex (vector) and the coin they used for each flip, dime or quarter (matrix). What is the best way to achieve this?

Below are two approaches I tried. Although they seem to work, they take too long to scale...

Let A be the array storing the coin flips, X the vector storing people's sex, and Y the matrix storing the coins used:

A <- array(sample(c("H","T"), size=n.i*n.t*n.k, replace=T), dim=c(n.i, n.t, n.k))
X <- as.logical(rbinom(n.i, 1, 0.49))
Y <- matrix(as.logical(rbinom(n.i*n.p, 1, 0.3)), nrow=n.i, ncol=n.k)

In my case n.i <- 10^5 n.t <- 10^2 n.k <-10

Approach 1 — Vectorized approach:

result <- matrix(0, nrow=n.t, ncol=4)
count <- matrix(0, nrow=n.i, ncol=n.t) 
heads <- A=="H"
for (x in 0:1) # male or female
 for (y in 0:1) # dime or quarter
 count <- 0
 for (k in 1:n.k) 
 count <- count + 1*(X==x & Y[,k]==y & heads[,,k])
 
 result[,1+x+2*(y-1)] <- colSums(count>0) 
 

Approach 2 — Expand X and Y to similar dimensions as A, fixing values along the k and t axis, eg: X <- array(X, dim=c(n.i, n.t, n.k)). Then use apply():

for (x in 0:1) 
 for (y in 0:1) 
 result[,1+x+2*(y-1)] <- apply(apply(X==x & Y==y & heads, 3, sum)>0, 2, sum) 
 
 

Anyone any better solutions?

data-management

share|improve this question

edited Mar 25 at 18:31

Reinier Meester

asked Mar 22 at 6:25

Reinier MeesterReinier Meester

12

add a comment | 

I want to summarize outcomes from a 3-dimensional array contingent on information from two other datasets. Say, the number of individuals i who after flipping k coins in year t (array) have at least 1 head, with results organized by the individual's sex (vector) and the coin they used for each flip, dime or quarter (matrix). What is the best way to achieve this?

Below are two approaches I tried. Although they seem to work, they take too long to scale...

Let A be the array storing the coin flips, X the vector storing people's sex, and Y the matrix storing the coins used:

A <- array(sample(c("H","T"), size=n.i*n.t*n.k, replace=T), dim=c(n.i, n.t, n.k))
X <- as.logical(rbinom(n.i, 1, 0.49))
Y <- matrix(as.logical(rbinom(n.i*n.p, 1, 0.3)), nrow=n.i, ncol=n.k)

In my case n.i <- 10^5 n.t <- 10^2 n.k <-10

Approach 1 — Vectorized approach:

result <- matrix(0, nrow=n.t, ncol=4)
count <- matrix(0, nrow=n.i, ncol=n.t) 
heads <- A=="H"
for (x in 0:1) # male or female
 for (y in 0:1) # dime or quarter
 count <- 0
 for (k in 1:n.k) 
 count <- count + 1*(X==x & Y[,k]==y & heads[,,k])
 
 result[,1+x+2*(y-1)] <- colSums(count>0) 
 

Approach 2 — Expand X and Y to similar dimensions as A, fixing values along the k and t axis, eg: X <- array(X, dim=c(n.i, n.t, n.k)). Then use apply():

for (x in 0:1) 
 for (y in 0:1) 
 result[,1+x+2*(y-1)] <- apply(apply(X==x & Y==y & heads, 3, sum)>0, 2, sum) 
 
 

Anyone any better solutions?

data-management

share|improve this question

edited Mar 25 at 18:31

Reinier Meester

asked Mar 22 at 6:25

Reinier MeesterReinier Meester

12

A <- array(sample(c("H","T"), size=n.i*n.t*n.k, replace=T), dim=c(n.i, n.t, n.k))
X <- as.logical(rbinom(n.i, 1, 0.49))
Y <- matrix(as.logical(rbinom(n.i*n.p, 1, 0.3)), nrow=n.i, ncol=n.k)

result <- matrix(0, nrow=n.t, ncol=4)
count <- matrix(0, nrow=n.i, ncol=n.t) 
heads <- A=="H"
for (x in 0:1) # male or female
 for (y in 0:1) # dime or quarter
 count <- 0
 for (k in 1:n.k) 
 count <- count + 1*(X==x & Y[,k]==y & heads[,,k])
 
 result[,1+x+2*(y-1)] <- colSums(count>0) 
 

for (x in 0:1) 
 for (y in 0:1) 
 result[,1+x+2*(y-1)] <- apply(apply(X==x & Y==y & heads, 3, sum)>0, 2, sum) 
 
 

share|improve this question

edited Mar 25 at 18:31

Reinier Meester

asked Mar 22 at 6:25

Reinier MeesterReinier Meester

12

share|improve this question

edited Mar 25 at 18:31

Reinier Meester

asked Mar 22 at 6:25

Reinier MeesterReinier Meester

12

asked Mar 22 at 6:25

Reinier MeesterReinier Meester

12

asked Mar 22 at 6:25

Reinier MeesterReinier Meester

12