I am trying to generate n random numbers whose sum is less than 1.
So I can't just run runif(3). But I can condition each iteration on the sum of all values generated up to that point.
The idea is to start an empty vector, v, and set up a loop such that for each iteration, i, a runif() is generated, but before it is accepted as an element of v, i.e. v[i] <- runif(), the test sum(v) < 1 is carried out, and while FALSE the last entry v[i] is finally accepted, BUT if TRUE, that is the sum is greater than 1, v[i] is tossed out of the vector, and the iteration i is repeated.
I am far from implementing this idea, but I would like to resolve it along the lines of something similar to what follows. It's not so much a practical problem, but more of an exercise to understand the syntax of loops in general:
n <- 4
v <- 0
for (i in 1:n){
rdom <- runif(1)
if((sum(v) + rdom) < 1) v[i] <- rdom
}
# keep trying before moving on to iteration i + 1???? i <- stays i?????
}
I have looked into while (actually I incorporated the while function in the title); however, I need the vector to have n elements, so I get stuck if I try something that basically tells R to add random uniform realizations as elements of the vector v while sum(v) < 1, because I can end up with less than n elements in v.
Here's a possible solution. It doesn't use while but the more generic repeat. I edited it to use a while and save a couple of lines.
set.seed(0)
n <- 4
v <- numeric(n)
i <- 0
while (i < n) {
ith <- runif(1)
if (sum(c(v, ith)) < 1) {
i <- i+1
v[i] <- ith
}
}
v
# [1] 0.89669720 0.06178627 0.01339033 0.02333120
Using a repeat block, you must check for the condition anyways, but, removing the growing problem, it would look very similar:
set.seed(0)
n <- 4
v <- numeric(n)
i <- 0
repeat {
ith <- runif(1)
if (sum(c(v, ith)) < 1) {
i <- i+1
v[i] <- ith
}
if (i == 4) break
}
If you really want to keep exactly the same procedure that you have posted (aka iteratively sample the n values one at a time from the standard uniform distribution, rejecting any samples that cause your sum to exceed 1), then the following code is mathematically equivalent, shorter, and more efficient:
samp <- function(n) {
v <- rep(0, n)
for (i in 1:n) {
v[i] <- runif(1, 0, 1-sum(v))
}
v
}
Basically, this code uses the mathematical fact that if the sum of the vector is currently sum(v), then sampling from the standard uniform distribution until you get a value no greater than 1-sum(v) is exactly equivalent to sampling in the uniform distribution from 0 to 1-sum(v). The advantage of using the latter approach is that it's much more efficient -- we don't need to keep rejecting samples and trying again, and can instead just sample once for each element.
To get a sense of the runtime differences, consider sampling 100 observations with n=10, comparing to a working implementation of the code from your post (copied from my other answer to this question):
OP <- function(n) {
v <- rep(0, n)
for (i in 1:n){
rdom <- runif(1)
while (sum(v) + rdom > 1) rdom <- runif(1)
v[i] <- rdom
}
v
}
set.seed(144)
system.time(samples.OP <- replicate(100, OP(10)))
# user system elapsed
# 261.937 1.641 265.805
system.time(samples.josliber <- replicate(100, samp(10)))
# user system elapsed
# 0.004 0.001 0.004
In this case, the new approach is approaching 100,000 times faster.
It sounds like you're trying to uniformly sample from a space of n variables where the following constraints hold:
x_1 + x_2 + ... + x_n <= 1
x_1 >= 0
x_2 >= 0
...
x_n >= 0
The "hit and run" algorithm is the mathematical machinery that enables you to do exactly this. In 2-dimensional space, the algorithm will sample uniformly from the following triangle, with each location in the shaded area being equally likely to be selected:
The algorithm is provided in R through the hitandrun package, which requires you to specify the linear inequalities that define the space through a constraint matrix, direction vector, and right-hand side vector:
library(hitandrun)
n <- 3
constr <- list(constr = rbind(rep(1, n), -diag(n)),
dir = c(rep("<=", n+1)),
rhs = c(1, rep(0, n)))
set.seed(144)
samples <- hitandrun(constr, n.samples=1000)
head(samples, 10)
# [,1] [,2] [,3]
# [1,] 0.28914690 0.01620488 0.42663224
# [2,] 0.65489979 0.28455231 0.00199671
# [3,] 0.23215115 0.00661661 0.63597912
# [4,] 0.29644234 0.06398131 0.60707269
# [5,] 0.58335047 0.13891392 0.06151205
# [6,] 0.09442808 0.30287832 0.55118290
# [7,] 0.51462261 0.44094683 0.02641638
# [8,] 0.38847794 0.15501252 0.31572793
# [9,] 0.52155055 0.09921046 0.13304728
# [10,] 0.70503030 0.03770875 0.14299089
Breaking down this code a bit, we generated the following constraint matrix:
constr
# $constr
# [,1] [,2] [,3]
# [1,] 1 1 1
# [2,] -1 0 0
# [3,] 0 -1 0
# [4,] 0 0 -1
#
# $dir
# [1] "<=" "<=" "<=" "<="
#
# $rhs
# [1] 1 0 0 0
Reading across the first line of constr$constr we have 1, 1, 1 which indicates "1*x1 + 1*x2 + 1*x3". The first element of constr$dir is <=, and the first element of constr$rhs is 1; putting it together we have x1 + x2 + x3 <= 1. From the second row of constr$constr we read -1, 0, 0 which indicates "-1*x1 + 0*x2 + 0*x3". The second element of constr$dir is <= and the second element of constr$rhs is 0; putting it together we have -x1 <= 0 which is the same as saying x1 >= 0. The similar non-negativity constraints follow in the remaining rows.
Note that the hit and run algorithm has the nice property of having the exact same distribution for each of the variables:
hist(samples[,1])
hist(samples[,2])
hist(samples[,3])
Meanwhile, the distribution of the samples from your procedure will be highly uneven, and as n increases this problem will get worse and worse.
OP <- function(n) {
v <- rep(0, n)
for (i in 1:n){
rdom <- runif(1)
while (sum(v) + rdom > 1) rdom <- runif(1)
v[i] <- rdom
}
v
}
samples.OP <- t(replicate(1000, OP(3)))
hist(samples.OP[,1])
hist(samples.OP[,2])
hist(samples.OP[,3])
An added advantage is that the hit-and-run algorithm appears faster -- I generated these 1000 replicates in 0.006 seconds on my computer with hit-and-run and it took 0.3 seconds using the modified code from the OP.
Here's how I would do it, without any loop, if or while:
set.seed(123)
x <- runif(1) # start with the sum that you want to obtain
n <- 4 # number of generated random numbers, can be chosen arbitrarily
y <- sort(runif(n-1,0,x)) # choose n-1 random points to cut the range [0:x]
z <- c(y[1],diff(y),x-y[n-1]) # result: determine the length of the segments
#> z
#[1] 0.11761257 0.10908627 0.02723712 0.03364156
#> sum(z)
#[1] 0.2875775
#> all.equal(sum(z),x)
#[1] TRUE
The advantage here is that you can determine exactly which sum you want to obtain and how many numbers n you want to generate for this. If you set, e.g., x <- 1 in the second line, the n random numbers stored in the vector z will add up to one.
Related
I just saw a YouTube video from Numberphile on the Yellowstone sequence (A098550). It's base on a sequence starting with 1 and 2, with subsequent terms generated by the rules:
no repeated terms
always pick the lowest integer
gcd(a_n, a_(n-1)) = 1
gcd(a_n, a_(n-2)) > 1
The first 15 terms would be: 1 2 3 4 9 8 15 14 5 6 25 12 35 16 7
A Q&D approach in R could be something like this, but understandably, this becomes very slow at attempts to make longer sequences. It also make some assumptions about the highest number that is possible within the sequence (as info: the sequence of 10,000 items never goes higher than 5000).
What can we do to make this faster?
library(DescTools)
a <- c(1, 2, 3)
p <- length(a)
# all natural numbers
all_ints <- 1:5000
for (n in p:1000) {
# rule 1 - remove all number that are in sequence already
next_a_set <- all_ints[which(!all_ints %in% a)]
# rule 3 - search the remaining set for numbers that have gcd == 1
next_a_option <- next_a_set[which(
sapply(
next_a_set,
function(x) GCD(a[n], x)
) == 1
)]
# rule 4 - search the remaining number for gcd > 1
next_a <- next_a_option[which(
sapply(
next_a_option,
function(x) GCD(a[n - 1], x)
) > 1
)]
# select the lowest
a <- c(a, min(next_a))
n <- n + 1
}
Here's a version that's about 20 times faster than yours, with comments about the changes:
# Set a to the final length from the start.
a <- c(1, 2, 3, rep(NA, 997))
p <- 3
# Define a vectorized gcd() function. We'll be testing
# lots of gcds at once. This uses the Euclidean algorithm.
gcd <- function(x, y) { # vectorized gcd
while (any(y != 0)) {
x1 <- ifelse(y == 0, x, y)
y <- ifelse(y == 0, 0, x %% y)
x <- x1
}
x
}
# Guess at a reasonably large vector to work from,
# but we'll grow it later if not big enough.
allnum <- 1:1000
# Keep a logical record of what has been used
used <- c(rep(TRUE, 3), rep(FALSE, length(allnum) - 3))
for (n in p:1000) {
# rule 1 - remove all number that are in sequence already
# nothing to do -- used already records that.
repeat {
# rule 3 - search the remaining set for numbers that have gcd == 1
keep <- !used & gcd(a[n], allnum) == 1
# rule 4 - search the remaining number for gcd > 1
keep <- keep & gcd(a[n-1], allnum) > 1
# If we found anything, break out of this loop
if (any(keep))
break
# Otherwise, make the set of possible values twice as big,
# and try again
allnum <- seq_len(2*length(allnum))
used <- c(used, rep(FALSE, length(used)))
}
# select the lowest
newval <- which.max(keep)
# Assign into the appropriate place
a[n+1] <- newval
# Record that it has been used
used[newval] <- TRUE
}
If you profile it, you'll see it spends most of its time in the gcd() function. You could probably make that a lot faster by redoing it in C or C++.
The biggest change here is pre-allocation and restricting the search to numbers that have not yet been used.
library(numbers)
N <- 5e3
a <- integer(N)
a[1:3] <- 1:3
b <- logical(N) # which numbers have been used already?
b[1:3] <- TRUE
NN <- 1:N
system.time({
for (n in 4:N) {
a1 <- a[n - 1L]
a2 <- a[n - 2L]
for (k in NN[!b]) {
if (GCD(k, a1) == 1L & GCD(k, a2) > 1L) {
a[n] <- k
b[k] <- TRUE
break
}
}
if (!a[n]) {
a <- a[1:(n - 1L)]
break
}
}
})
#> user system elapsed
#> 1.28 0.00 1.28
length(a)
#> [1] 1137
For a fast C++ algorithm, see here.
I have two large sparse matrices (about 41,000 x 55,000 in size). The density of nonzero elements is around 10%. They both have the same row index and column index for nonzero elements.
I now want to modify the values in the first sparse matrix if values in the second matrix are below a certain threshold.
library(Matrix)
# Generating the example matrices.
set.seed(42)
# Rows with values.
i <- sample(1:41000, 227000000, replace = TRUE)
# Columns with values.
j <- sample(1:55000, 227000000, replace = TRUE)
# Values for the first matrix.
x1 <- runif(227000000)
# Values for the second matrix.
x2 <- sample(1:3, 227000000, replace = TRUE)
# Constructing the matrices.
m1 <- sparseMatrix(i = i, j = j, x = x1)
m2 <- sparseMatrix(i = i, j = j, x = x2)
I now get the rows, columns and values from the first matrix in a new matrix. This way, I can simply subset them and only the ones I am interested in remain.
# Getting the positions and values from the matrices.
position_matrix_from_m1 <- rbind(i = m1#i, j = summary(m1)$j, x = m1#x)
position_matrix_from_m2 <- rbind(i = m2#i, j = summary(m2)$j, x = m2#x)
# Subsetting to get the elements of interest.
position_matrix_from_m1 <- position_matrix_from_m1[,position_matrix_from_m1[3,] > 0 & position_matrix_from_m1[3,] < 0.05]
# We add 1 to the values, since the sparse matrix is 0-based.
position_matrix_from_m1[1,] <- position_matrix_from_m1[1,] + 1
position_matrix_from_m1[2,] <- position_matrix_from_m1[2,] + 1
Now I am getting into trouble. Overwriting the values in the second matrix takes too long. I let it run for several hours and it did not finish.
# This takes hours.
m2[position_matrix_from_m1[1,], position_matrix_from_m1[2,]] <- 1
m1[position_matrix_from_m1[1,], position_matrix_from_m1[2,]] <- 0
I thought about pasting the row and column information together. Then I have a unique identifier for each value. This also takes too long and is probably just very bad practice.
# We would get the unique identifiers after the subsetting.
m1_identifiers <- paste0(position_matrix_from_m1[1,], "_", position_matrix_from_m1[2,])
m2_identifiers <- paste0(position_matrix_from_m2[1,], "_", position_matrix_from_m2[2,])
# Now, I could use which and get the position of the values I want to change.
# This also uses to much memory.
m2_identifiers_of_interest <- which(m2_identifiers %in% m1_identifiers)
# Then I would modify the x values in the position_matrix_from_m2 matrix and overwrite m2#x in the sparse matrix object.
Is there a fundamental error in my approach? What should I do to run this efficiently?
Is there a fundamental error in my approach?
Yes. Here it is.
# This takes hours.
m2[position_matrix_from_m1[1,], position_matrix_from_m1[2,]] <- 1
m1[position_matrix_from_m1[1,], position_matrix_from_m1[2,]] <- 0
Syntax as mat[rn, cn] (whether mat is a dense or sparse matrix) is selecting all rows in rn and all columns in cn. So you get a length(rn) x length(cn) matrix. Here is a small example:
A <- matrix(1:9, 3, 3)
# [,1] [,2] [,3]
#[1,] 1 4 7
#[2,] 2 5 8
#[3,] 3 6 9
rn <- 1:2
cn <- 2:3
A[rn, cn]
# [,1] [,2]
#[1,] 4 7
#[2,] 5 8
What you intend to do is to select (rc[1], cn[1]), (rc[2], cn[2]) ..., only. The correct syntax is then mat[cbind(rn, cn)]. Here is a demo:
A[cbind(rn, cn)]
#[1] 4 8
So you need to fix your code to:
m2[cbind(position_matrix_from_m1[1,], position_matrix_from_m1[2,])] <- 1
m1[cbind(position_matrix_from_m1[1,], position_matrix_from_m1[2,])] <- 0
Oh wait... Based on your construction of position_matrix_from_m1, this is just
ij <- t(position_matrix_from_m1[1:2, ])
m2[ij] <- 1
m1[ij] <- 0
Now, let me explain how you can do better. You have underused summary(). It returns a 3-column data frame, giving (i, j, x) triplet, where both i and j are index starting from 1. You could have worked with this nice output directly, as follows:
# Getting (i, j, x) triplet (stored as a data.frame) for both `m1` and `m2`
position_matrix_from_m1 <- summary(m1)
# you never seem to use `position_matrix_from_m2` so I skip it
# Subsetting to get the elements of interest.
position_matrix_from_m1 <- subset(position_matrix_from_m1, x > 0 & x < 0.05)
Now you can do:
ij <- as.matrix(position_matrix_from_m1[, 1:2])
m2[ij] <- 1
m1[ij] <- 0
Is there a even better solution? Yes! Note that nonzero elements in m1 and m2 are located in the same positions. So basically, you just need to change m2#x according to m1#x.
ind <- m1#x > 0 & m1#x < 0.05
m2#x[ind] <- 1
m1#x[ind] <- 0
A complete R session
I don't have enough RAM to create your large matrix, so I reduced your problem size a little bit for testing. Everything worked smoothly.
library(Matrix)
# Generating the example matrices.
set.seed(42)
## reduce problem size to what my laptop can bear with
squeeze <- 0.1
# Rows with values.
i <- sample(1:(41000 * squeeze), 227000000 * squeeze ^ 2, replace = TRUE)
# Columns with values.
j <- sample(1:(55000 * squeeze), 227000000 * squeeze ^ 2, replace = TRUE)
# Values for the first matrix.
x1 <- runif(227000000 * squeeze ^ 2)
# Values for the second matrix.
x2 <- sample(1:3, 227000000 * squeeze ^ 2, replace = TRUE)
# Constructing the matrices.
m1 <- sparseMatrix(i = i, j = j, x = x1)
m2 <- sparseMatrix(i = i, j = j, x = x2)
## give me more usable RAM
rm(i, j, x1, x2)
##
## fix to your code
##
m1a <- m1
m2a <- m2
# Getting (i, j, x) triplet (stored as a data.frame) for both `m1` and `m2`
position_matrix_from_m1 <- summary(m1)
# Subsetting to get the elements of interest.
position_matrix_from_m1 <- subset(position_matrix_from_m1, x > 0 & x < 0.05)
ij <- as.matrix(position_matrix_from_m1[, 1:2])
m2a[ij] <- 1
m1a[ij] <- 0
##
## the best solution
##
m1b <- m1
m2b <- m2
ind <- m1#x > 0 & m1#x < 0.05
m2b#x[ind] <- 1
m1b#x[ind] <- 0
##
## they are identical
##
all.equal(m1a, m1b)
#[1] TRUE
all.equal(m2a, m2b)
#[1] TRUE
Caveat:
I know that some people may propose
m1c <- m1
m2c <- m2
logi <- m1 > 0 & m1 < 0.05
m2c[logi] <- 1
m1c[logi] <- 0
It looks completely natural in R's syntax. But trust me, it is extremely slow for large matrices.
Say I have a vector of random numbers, I can order them lowest to highest:
set.seed(1)
x <- runif(20)
v <- x[order(x)]
Now, say I want to order them but with some degree of noise.
I can randomly move elements like this:
z <-sample(1:20,2)
replace(v, z, v[rev(z)])
but this doesn't necessarily move closely related values. I could be equally likely to randomly switch the 1st and 20th values as the 5th and 6th. I would like to have some control over the switching, so I can switch more closely related values.
Ideally, I would be able to reorder the vector to have a specific Spearman's correlation. Say rather than the Spearman correlation of rank order being 1 when they are perfectly ordered, is there a way to reorder that same vector of numbers to have e.g. a Spearman's correlation of 0.5 ?
What if you added some noise to their rankings. This will makes sure values don't get moved too far away from the starting point. For example
set.seed(1)
N <- 50
D <- 3 # controls how far things can move
x <- runif(N)
v <- x[vx <- order(rank(x) + runif(N, -D, D))]
z <- x[order(x)]
layout(matrix(c(1,3,2,3), nrow=2))
plot(v, main ="Ordered")
plot(z, main ="Mixed")
plot(v, z, xlab="ordered", ylab="mixed"); abline(0,1)
I don't think I have completely understood your question but here's a start. I am simply recursively swapping random consecutive values of the sorted vector. You can control the amount of swapping with n_swaps argument. -
noisy_sort <- function(x, n_swaps) {
sorted_x <- sort(x)
indices <- sample(seq_along(x[-1]), n_swaps)
for(i in indices) {
sorted_x[c(i, i+1)] <- sorted_x[c(i+1, i)]
}
sorted_x
}
set.seed(1)
x <- runif(20)
result <- noisy_sort(x, 3)
order(result)
[1] 1 2 3 5 4 6 7 8 9 10 11 13 12 14 15 16 17 19 18 20
^ ^ ^ ^ ^ ^
Here is a very rudimentary algo.
Using Spearman correlation for distinct ranks, you can back out the desired sum of squared difference (SSE) between ranks. Then, using a Markov Chain Monte Carlo (MCMC) approach, you sample a pair of indices to swap and transit to the new vector with swapped elements if it improves the SSE towards desired score.
I used the number of iterations as the stopping criteria. You can change the condition so that it meets a target tolerance level.
set.seed(1)
n <- 20
x <- runif(n)
v <- sort(x)
calc_exp_sse <- function(rho, N) {
(1 - rho) * N * (N^2 - 1) / 6
}
exp_sse <- calc_exp_sse(0.5, n)
ord <- 1:n
vec <- ord
for (i in 1:1000) {
swap <- vec
swid <- sample(n, 2L)
swap[swid] <- swap[c(swid[2L], swid[1L])]
if (abs(exp_sse - sum((ord-swap)^2)) < abs(exp_sse - sum((ord-vec)^2))) {
vec <- swap
}
}
vec
cor(vec, ord, method="spearman")
#[1] 0.5007519
cor(v, v[vec], method="spearman")
#[1] 0.5007519
I am writing a function to perform bit inversion for each row of a binary matrix which depends on a predefined n value. The n value will determine the number of 1 bits for each row of the matrix.
set.seed(123)
## generate a random 5 by 10 binary matrix
init <- t(replicate(5, {i <- sample(3:6, 1); sample(c(rep(1, i), rep(0, 10 - i)))}))
n <- 3
## init_1 is a used to explain my problem (single row matrix)
init_1 <- t(replicate(1, {i <- sample(3:6, 1); sample(c(rep(1, i), rep(0, 10 - i)))}))
The bit_inversion function does this few things:
If the selected row has number of 1's lesser than n, then it randomly select a few indices (difference) and invert them. (0 to 1)
Else if the selected row has number of 1's greater than n, then it randomly select a few indices (difference) and invert them. (1 to 0)
Else do nothing (when the row has number of 1's equals to n.)
Below is the function I implemented:
bit_inversion<- function(pop){
for(i in 1:nrow(pop)){
difference <- abs(sum(pop[i,]) - n)
## checking condition where there are more bits being turned on than n
if(sum(pop[i,]) > n){
## determine position of 1's
bit_position_1 <- sample(which(pop[i,]==1), difference)
## bit inversion
for(j in 1:length(bit_position_1)){
pop[bit_position_1[j]] <- abs(pop[i,][bit_position_1[j]] - 1)
}
}
else if (sum(pop[i,]) < n){
## determine position of 0's
bit_position_0 <- sample(which(pop[i,]==0), difference)
## bit inversion
for(j in 1:length(bit_position_0)){
pop[bit_position_0[j]] <- abs(pop[bit_position_0[j]] - 1)
}
}
}
return(pop)
}
Outcome:
call <- bit_inversion(init)
> rowSums(call) ## suppose to be all 3
[1] 3 4 5 4 3
But when using init_1 (a single row matrix), the function seems to work fine.
Outcome:
call_1 <- bit_inversion(init_1)
> rowSums(call)
[1] 3
Is there a mistake in my for and if...else loop?
Change the line in 'j' for loop
pop[bit_position_1[j]] <- abs(pop[i,][bit_position_1[j]] - 1)
into
pop[i,bit_position_1[j]] <- abs(pop[i,][bit_position_1[j]] - 1)
You forgot the row index.
And, here is a more compact version of your for loop:
for(i in 1:nrow(pop)){
difference <- abs(sum(pop[i,]) - n)
logi <- sum(pop[i,]) > n
pop[i,sample(which(pop[i,]==logi), difference)] <- !logi
}
We have a big for loop in R for simulating various data where for some iterations the data generate in such a way that a quantity comes 0 inside the loop, which is not desirable and we should skip that step of data generation. But at the same time we also need to increase the number of iterations by one step because of such skip, otherwise we will have fewer observations than required.
For example, while running the following code, we get z=0 in iteration 1, 8 and 9.
rm(list=ls())
n <- 10
z <- NULL
for(i in 1:n){
set.seed(i)
a <- rbinom(1,1,0.5)
b <- rbinom(1,1,0.5)
z[i] <- a+b
}
z
[1] 0 1 1 1 1 2 1 0 0 1
We desire to skip these steps so that we do not have any z=0 but we also want a vector z of length 10. It may be done in many ways. But what I particularly want to see is how we can stop the iteration and skip the current step when z=0 is encountered and go to the next step, ultimately obtaining 10 observations for z.
Normally we do this via a while loop, as the number of iterations required is unknown beforehand.
n <- 10L
z <- integer(n)
m <- 1L; i <- 0L
while (m <= n) {
set.seed(i)
z_i <- sum(rbinom(2L, 1, 0.5))
if (z_i > 0L) {z[m] <- z_i; m <- m + 1L}
i <- i + 1L
}
Output:
z
# [1] 1 1 1 1 1 2 1 1 1 1
i
# [1] 14
So we sample 14 times, 4 of which are 0 and the rest 10 are retained.
More efficient vectorized method
set.seed(0)
n <- 10L
z <- rbinom(n, 1, 0.5) + rbinom(n, 1, 0.5)
m <- length(z <- z[z > 0L]) ## filtered samples
p <- m / n ## estimated success probability
k <- round(1.5 * (n - m) / p) ## further number of samples to ensure successful (n - m) non-zero samples
z_more <- rbinom(k, 1, 0.5) + rbinom(k, 1, 0.5)
z <- c(z, z_more[which(z_more > 0)[seq_len(n - m)]])
Some probability theory of geometric distribution has been used here. Initially we sample n samples, m of which are retained. So the estimated probability of success in accepting samples is p <- m/n. According to theory of Geometric distribution, on average, we need at least 1/p samples to observe a success. Therefore, we should at least sample (n-m)/p more times to expect (n-m) success. The 1.5 is just an inflation factor. By sampling 1.5 times more samples we hopefully can ensure (n-m) success.
According to Law of large numbers, the estimate of p is more precise when n is large. Therefore, this approach is stable for large n.
If you feel that 1.5 is not large enough, use 2 or 3. But my feeling is that it is sufficient.