Julia Convex.jl unbounded error - julia

I'm trying to calculate the spectral norm of A, this seems straight forward but the solver is telling me it's unbounded, which doesn't make sense, since y must have unit norm.
using Convex
using SCS
set_default_solver(SCSSolver(verbose=0))
A = [1 2; 3 4]
y = Variable(2)
expr = norm(A*y, 2)
constr = norm(y, 2) == 1
problem = maximize(expr, constr)
solve!(problem)
Am I missing something?
Edit: Removing solve!(problem) from the code (thus just setting things up), results in a warning that the problem is not DCP (disciplined convex programming) compliant, yet since this is just calculating the spectral norm of A, it should be convex.

Related

Initial state starts at y(1), how to go backwards to find y(0)? [duplicate]

I would like to solve a differential equation in R (with deSolve?) for which I do not have the initial condition, but only the final condition of the state variable. How can this be done?
The typical code is: ode(times, y, parameters, function ...) where y is the initial condition and function defines the differential equation.
Are your equations time reversible, that is, can you change your differential equations so they run backward in time? Most typically this will just mean reversing the sign of the gradient. For example, for a simple exponential growth model with rate r (gradient of x = r*x) then flipping the sign makes the gradient -r*x and generates exponential decay rather than exponential growth.
If so, all you have to do is use your final condition(s) as your initial condition(s), change the signs of the gradients, and you're done.
As suggested by #LutzLehmann, there's an even easier answer: ode can handle negative time steps, so just enter your time vector as (t_end, 0). Here's an example, using f'(x) = r*x (i.e. exponential growth). If f(1) = 3, r=1, and we want the value at t=0, analytically we would say:
x(T) = x(0) * exp(r*T)
x(0) = x(T) * exp(-r*T)
= 3 * exp(-1*1)
= 1.103638
Now let's try it in R:
library(deSolve)
g <- function(t, y, parms) { list(parms*y) }
res <- ode(3, times = c(1, 0), func = g, parms = 1)
print(res)
## time 1
## 1 1 3.000000
## 2 0 1.103639
I initially misread your question as stating that you knew both the initial and final conditions. This type of problem is called a boundary value problem and requires a separate class of numerical algorithms from standard (more elementary) initial-value problems.
library(sos)
findFn("{boundary value problem}")
tells us that there are several R packages on CRAN (bvpSolve looks the most promising) for solving these kinds of problems.
Given a differential equation
y'(t) = F(t,y(t))
over the interval [t0,tf] where y(tf)=yf is given as initial condition, one can transform this into the standard form by considering
x(s) = y(tf - s)
==> x'(s) = - y'(tf-s) = - F( tf-s, y(tf-s) )
x'(s) = - F( tf-s, x(s) )
now with
x(0) = x0 = yf.
This should be easy to code using wrapper functions and in the end some list reversal to get from x to y.
Some ODE solvers also allow negative step sizes, so that one can simply give the times for the construction of y in the descending order tf to t0 without using some intermediary x.

optiSolve package in r

I'm trying to maximize the portfolio return subject to 5 constraints:
1.- a certain level of portfolio risk
2.- the same above but oposite sign (I need that the risk to be exactly that number)
3.- the sum of weights have to be 1
4.- all the weights must be greater or equal to cero
5.- all the weights must be at most one
I'm using the optiSolve package because I didn't find any other package that allow me to write this problem (or al least that I understood how to use it).
I have three big problems here, the first is that the resulting weights vector sum more than 1 and the second problem is that I can't declare t(w) %*% varcov_matrix %*% w == 0 in the quadratic constraint because it only allows for "<=" and finally I don't know how to put a constraint to get only positives weights
vector_de_retornos <- rnorm(5)
matriz_de_varcov <- matrix(rnorm(25), ncol = 5)
library(optiSolve)
restriccion1 <- quadcon(Q = matriz_de_varcov, dir = "<=", val = 0.04237972)
restriccion1_neg <- quadcon(Q = -matriz_de_varcov, dir = "<=",
val = -mean(limite_inf, limite_sup))
restriccion2 <- lincon(t(vector_de_retornos),
d=rep(0, nrow(t(vector_de_retornos))),
dir=rep("==",nrow(t(vector_de_retornos))),
val = rep(1, nrow(t(vector_de_retornos))),
id=1:ncol(t(vector_de_retornos)),
name = nrow(t(vector_de_retornos)))
restriccion_nonnegativa <- lbcon(rep(0,length(vector_de_retornos)))
restriccion_positiva <- ubcon(rep(1,length(vector_de_retornos)))
funcion_lineal <- linfun(vector_de_retornos, name = "lin.fun")
funcion_obj <- cop(funcion_lineal, max = T, ub = restriccion_positiva,
lc = restriccion2, lb = restriccion_nonnegativa, restriccion1,
restriccion1_neg)
porfavor_funciona <- solvecop(funcion_obj, solver = "alabama")
> porfavor_funciona$x
1 2 3 4 5
-3.243313e-09 -4.709673e-09 9.741379e-01 3.689040e-01 -1.685290e-09
> sum(porfavor_funciona$x)
[1] 1.343042
Someone knows how to solve this maximization problem with all the constraints mentioned before or tell me what I'm doing wrong? I'll really appreciate that, because the result seems like is not taking into account the constraints. Thanks!
Your restriccion2 makes the weighted sum of x is 1, if you also want to ensure the regular sum of x is 1, you can modify the constraint as follows:
restriccion2 <- lincon(rbind(t(vector_de_retornos),
# make a second row of coefficients in the A matrix
t(rep(1,length(vector_de_retornos)))),
d=rep(0,2), # the scalar value for both constraints is 0
dir=rep('==',2), # the direction for both constraints is '=='
val=rep(1,2), # the rhs value for both constraints is 1
id=1:ncol(t(vector_de_retornos)), # the number of columns is the same as before
name= 1:2)
If you only want the regular sum to be 1 and not the weighted sum you can replace your first parameter in the lincon function as you've defined it to be t(rep(1,length(vector_de_retornos))) and that will just constrain the regular sum of x to be 1.
To make an inequality constraint using only inequalities you need the same constraint twice but with opposite signs on the coefficients and right hand side values between the two (for example: 2x <= 4 and -2x <= -4 combines to make the constraint 2*x == 4). In your edit above, you provide a different value to the val parameter so these two constraints won't combine to make the equality constraint unless they match except for opposite signs as below.
restriccion1_neg <- quadcon(Q = -matriz_de_varcov, dir = "<=", val = -0.04237972)
I'm not certain because I can't find precision information in the package documentation, but those "negative" values in the x vector are probably due to rounding. They are so small and are effectively 0 so I think the non-negativity constraint is functioning properly.
restriccion_nonnegativa <- lbcon(rep(0,length(vector_de_retornos)))
A constraint of the form
x'Qx = a
is non-convex. (More general: any nonlinear equality constraint is non-convex). Non-convex problems are much more difficult to solve than convex ones and require specialized, global solvers. For convex problems, there are quite a few solvers available. This is not the case for non-convex problems. Most portfolio models are formulated as convex QP (quadratic programming i.e. risk -- the quadratic term -- is in the objective) or convex QCP/SOCP problems (quadratic terms in the constraints, but in a convex fashion). So, the constraint
x'Qx <= a
is easy (convex), as long as Q is positive-semi definite. Rewriting x'Qx=a as
x'Qx <= a
-x'Qx <= -a
unfortunately does not make the non-convexity go away, as -Q is not PSD. If we are maximizing return, we usually only use x'Qx <= a to limit the risk and forget about the >= part. Even more popular is to put both the return and the risk in the objective (that is the standard mean-variable portfolio model).
A possible solver for solving non-convex quadratic problems under R is Gurobi.

How do I minimize a linear least squares function in R?

I'm reading Deep Learning by Goodfellow et al. and am trying to implement gradient descent as shown in Section 4.5 Example: Linear Least Squares. This is page 92 in the hard copy of the book.
The algorithm can be viewed in detail at https://www.deeplearningbook.org/contents/numerical.html with R implementation of linear least squares on page 94.
I've tried implementing in R, and the algorithm as implemented converges on a vector, but this vector does not seem to minimize the least squares function as required. Adding epsilon to the vector in question frequently produces a "minimum" less than the minimum outputted by my program.
options(digits = 15)
dim_square = 2 ### set dimension of square matrix
# Generate random vector, random matrix, and
set.seed(1234)
A = matrix(nrow = dim_square, ncol = dim_square, byrow = T, rlnorm(dim_square ^ 2)/10)
b = rep(rnorm(1), dim_square)
# having fixed A & B, select X randomly
x = rnorm(dim_square) # vector length of dim_square--supposed to be arbitrary
f = function(x, A, b){
total_vector = A %*% x + b # this is the function that we want to minimize
total = 0.5 * sum(abs(total_vector) ^ 2) # L2 norm squared
return(total)
}
f(x,A,b)
# how close do we want to get?
epsilon = 0.1
delta = 0.01
value = (t(A) %*% A) %*% x - t(A) %*% b
L2_norm = (sum(abs(value) ^ 2)) ^ 0.5
steps = vector()
while(L2_norm > delta){
x = x - epsilon * value
value = (t(A) %*% A) %*% x - t(A) %*% b
L2_norm = (sum(abs(value) ^ 2)) ^ 0.5
print(L2_norm)
}
minimum = f(x, A, b)
minimum
minimum_minus = f(x - 0.5*epsilon, A, b)
minimum_minus # less than the minimum found by gradient descent! Why?
On page 94 of the pdf appearing at https://www.deeplearningbook.org/contents/numerical.html
I am trying to find the values of the vector x such that f(x) is minimized. However, as demonstrated by the minimum in my code, and minimum_minus, minimum is not the actual minimum, as it exceeds minimum minus.
Any idea what the problem might be?
Original Problem
Finding the value of x such that the quantity Ax - b is minimized is equivalent to finding the value of x such that Ax - b = 0, or x = (A^-1)*b. This is because the L2 norm is the euclidean norm, more commonly known as the distance formula. By definition, distance cannot be negative, making its minimum identically zero.
This algorithm, as implemented, actually comes quite close to estimating x. However, because of recursive subtraction and rounding one quickly runs into the problem of underflow, resulting in massive oscillation, below:
Value of L2 Norm as a function of step size
Above algorithm vs. solve function in R
Above we have the results of A %% x followed by A %% min_x, with x estimated by the implemented algorithm and min_x estimated by the solve function in R.
The problem of underflow, well known to those familiar with numerical analysis, is probably best tackled by the programmers of lower-level libraries best equipped to tackle it.
To summarize, the algorithm appears to work as implemented. Important to note, however, is that not every function will have a minimum (think of a straight line), and also be aware that this algorithm should only be able to find a local, as opposed to a global minimum.

how to specify final value (rather than initial value) for solving differential equations

I would like to solve a differential equation in R (with deSolve?) for which I do not have the initial condition, but only the final condition of the state variable. How can this be done?
The typical code is: ode(times, y, parameters, function ...) where y is the initial condition and function defines the differential equation.
Are your equations time reversible, that is, can you change your differential equations so they run backward in time? Most typically this will just mean reversing the sign of the gradient. For example, for a simple exponential growth model with rate r (gradient of x = r*x) then flipping the sign makes the gradient -r*x and generates exponential decay rather than exponential growth.
If so, all you have to do is use your final condition(s) as your initial condition(s), change the signs of the gradients, and you're done.
As suggested by #LutzLehmann, there's an even easier answer: ode can handle negative time steps, so just enter your time vector as (t_end, 0). Here's an example, using f'(x) = r*x (i.e. exponential growth). If f(1) = 3, r=1, and we want the value at t=0, analytically we would say:
x(T) = x(0) * exp(r*T)
x(0) = x(T) * exp(-r*T)
= 3 * exp(-1*1)
= 1.103638
Now let's try it in R:
library(deSolve)
g <- function(t, y, parms) { list(parms*y) }
res <- ode(3, times = c(1, 0), func = g, parms = 1)
print(res)
## time 1
## 1 1 3.000000
## 2 0 1.103639
I initially misread your question as stating that you knew both the initial and final conditions. This type of problem is called a boundary value problem and requires a separate class of numerical algorithms from standard (more elementary) initial-value problems.
library(sos)
findFn("{boundary value problem}")
tells us that there are several R packages on CRAN (bvpSolve looks the most promising) for solving these kinds of problems.
Given a differential equation
y'(t) = F(t,y(t))
over the interval [t0,tf] where y(tf)=yf is given as initial condition, one can transform this into the standard form by considering
x(s) = y(tf - s)
==> x'(s) = - y'(tf-s) = - F( tf-s, y(tf-s) )
x'(s) = - F( tf-s, x(s) )
now with
x(0) = x0 = yf.
This should be easy to code using wrapper functions and in the end some list reversal to get from x to y.
Some ODE solvers also allow negative step sizes, so that one can simply give the times for the construction of y in the descending order tf to t0 without using some intermediary x.

Errors when attempting constrained optimisation using optim()

I have been using the Excel solver to handle the following problem
solve for a b and c in the equation:
y = a*b*c*x/((1 - c*x)(1 - c*x + b*c*x))
subject to the constraints
0 < a < 100
0 < b < 100
0 < c < 100
f(x[1]) < 10
f(x[2]) > 20
f(x[3]) < 40
where I have about 10 (x,y) value pairs. I minimize the sum of abs(y - f(x)). And I can constrain both the coefficients and the range of values for the result of my function at each x.
I tried nls (without trying to impose the constraints) and while Excel provided estimates for almost any starting values I cared to provide, nls almost never returned an answer.
I switched to using optim, but I'm having trouble applying the constraints.
This is where I have gotten so far-
best = function(p,x,y){sum(abs(y - p[1]*p[2]*p[3]*x/((1 - p[3]*x)*(1 - p[3]*x + p[2]*p[3]*x))))}
p = c(1,1,1)
x = c(.1,.5,.9)
y = c(5,26,35)
optim(p,best,x=x,y=y)
I did this to add the first set of constraints-
optim(p,best,x=x,y=y,method="L-BFGS-B",lower=c(0,0,0),upper=c(100,100,100))
I get the error ""ERROR: ABNORMAL_TERMINATION_IN_LNSRCH"
and end up with a higher value of the error ($value). So it seems like I am doing something wrong. I couldn't figure out how to apply my other set of constraints at all.
Could someone provide me a basic idea how to solve this problem that a non-statistician can understand? I looked at a lot of posts and looked in a few R books. The R books stopped at the simplest use of optim.
The absolute value introduces a singularity:
you may want to use a square instead,
especially for gradient-based methods (such as L-BFGS).
The denominator of your function can be zero.
The fact that the parameters appear in products
and that you allow them to be (arbitrarily close to) zero
can also cause problems.
You can try with other optimizers
(complete list on the optimization task view),
until you find one for which the optimization converges.
x0 <- c(.1,.5,.9)
y0 <- c(5,26,35)
p <- c(1,1,1)
lower <- 0*p
upper <- 100 + lower
f <- function(p,x=x0,y=y0) sum(
(
y - p[1]*p[2]*p[3]*x / ( (1 - p[3]*x)*(1 - p[3]*x + p[2]*p[3]*x) )
)^2
)
library(dfoptim)
nmkb(p, f, lower=lower, upper=upper) # Converges
library(Rvmmin)
Rvmmin(p, f, lower=lower, upper=upper) # Does not converge
library(DEoptim)
DEoptim(f, lower, upper) # Does not converge
library(NMOF)
PSopt(f, list(min=lower, max=upper))[c("xbest", "OFvalue")] # Does not really converge
DEopt(f, list(min=lower, max=upper))[c("xbest", "OFvalue")] # Does not really converge
library(minqa)
bobyqa(p, f, lower, upper) # Does not really converge
As a last resort, you can always use a grid search.
library(NMOF)
r <- gridSearch( f,
lapply(seq_along(p), function(i) seq(lower[i],upper[i],length=200))
)

Resources