lequant40 / portfolio_allocation_js Goto Github PK

X ref. from #7, to better separate issues.

Thanks Roman. That did the trick! Now I'm running into a different issue and would greatly appreciate any help:

I'm using the function below I created in Google Sheets, but am getting a #ERROR! that says "Error: unsupported problem detected (line 5838)". My goal is to calculate mean variance weights given an expected return stream, a covariance matrix, a target volatility, min weights, and max weights:

function computeWts(rets, cv, targetVol, minWts, maxWts) {
return (PortfolioAllocation.meanVarianceOptimizationWeights(rets, cv, {constraints: {optimizationMethod: 'maximumTargetVolatility', maxVolatility: targetVol, minWeights: minWts, maxWeights: maxWts}}));
}

The arguments I'm passing are below (and are all in cells in Google Sheets):

rets:
4.7%
5.0%
6.1%
4.0%
5.4%
2.0%
2.2%
2.4%
2.4%
3.6%

cv:
2.3% | 2.5% | 3.3% | 1.7% | 2.3% | -0.4% | -0.3% | -0.3% | 0.1% | 1.7%
2.5% | 2.9% | 3.8% | 1.9% | 2.5% | -0.4% | -0.3% | -0.3% | 0.2% | 1.9%
3.3% | 3.8% | 5.3% | 2.3% | 3.1% | -0.5% | -0.3% | -0.4% | 0.3% | 2.9%
1.7% | 1.9% | 2.3% | 1.4% | 1.8% | -0.3% | -0.3% | -0.3% | 0.0% | 1.1%
2.3% | 2.5% | 3.1% | 1.8% | 4.0% | -0.1% | 0.0% | 0.2% | 0.7% | 1.8%
-0.4% | -0.4% | -0.5% | -0.3% | -0.1% | 0.2% | 0.2% | 0.2% | 0.2% | -0.1%
-0.3% | -0.3% | -0.3% | -0.3% | 0.0% | 0.2% | 0.4% | 0.2% | 0.3% | 0.0%
-0.3% | -0.3% | -0.4% | -0.3% | 0.2% | 0.2% | 0.2% | 0.3% | 0.2% | 0.0%
0.1% | 0.2% | 0.3% | 0.0% | 0.7% | 0.2% | 0.3% | 0.2% | 0.5% | 0.4%
1.7% | 1.9% | 2.9% | 1.1% | 1.8% | -0.1% | 0.0% | 0.0% | 0.4% | 2.0%

targetVol:
5.5%

minWts:
3%
3%
3%
0%
0%
0%
0%
3%
0%
0%

maxWts:
35%
35%
35%
35%
35%
35%
35%
35%
35%
35%

Originally posted by @grantackerman1 in #7 (comment)

Describe the bug
When min/max weights constraints are provided in the meanVarianceOptimizationWeights portfolio allocation method, using the optimization method maximumTargetVolatility, an internal error is produced with the input data below.

	var covMat = [[0.0004246616877,-0.00005520852069,0.0003954256465,0.00000001152437917,0.0001590470407,0.0002580332644,0.0003335881244,0.0003605784739],[-0.00005520852069,0.00002480135555,-0.00005059666822,-0.00000001082496581,-0.00001673167975,-0.00003073486553,-0.00003900214181,-0.00004548279667],[0.0003954256465,-0.00005059666822,0.0003976604822,-0.000000008144301394,0.0001740534739,0.0002593777442,0.0003420673729,0.0003593307083],[0.00000001152437917,-0.00000001082496581,-0.000000008144301394,0.00000001790999881,-0.00000008021468402,-0.00000006461917657,-0.00000007037516421,0.00000004991265269],[0.0001590470407,-0.00001673167975,0.0001740534739,-0.00000008021468402,0.000110482962,0.0001154225601,0.000157800705,0.0001589926655],[0.0002580332644,-0.00003073486553,0.0002593777442,-0.00000006461917657,0.0001154225601,0.0002185122484,0.0002506289478,0.0002558291246],[0.0003335881244,-0.00003900214181,0.0003420673729,-0.00000007037516421,0.000157800705,0.0002506289478,0.0003278326867,0.0003411302813],[0.0003605784739,-0.00004548279667,0.0003593307083,0.00000004991265269,0.0001589926655,0.0002558291246,0.0003411302813,0.0004078706675]];

	var returns = [0.02836270809, 0.01289860823, 0.003013461519, 0.0009824011995, -0.002791817978, -0.003228292398, -0.01390104713, -0.01415404063];

	var minWeights = [0, 0, 0, 0, 0, 0, 0, 0];
	var maxWeights = [0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 1, 1];

	var maxVolatility = 0.05/Math.sqrt(252);

	PortfolioAllocation.meanVarianceOptimizationWeights(returns, covMat, { optimizationMethod: 'maximumTargetVolatility', constraints: {maxVolatility: maxVolatility, minWeights: minWeights, maxWeights: maxWeights}});

To Reproduce
C.f. above.

Expected behavior
No internal error, but proper portfolio weights should be computed !

Link to Google Sheets / On-line npm code
N/A

Additional context
N/A

Describe the bug
The meanVarianceOptimizationWeights method returns "Maximum number of iterations reached", whatever the number of iterations set, when using the following inputs:

	var covMat = [[0.04902254557, 0.04255604021, 0.04501327517, 0.04330779376, 0.03019691712, 0.02548665503, -0.01073017105, -0.0004006905689],
				[0.04682328959, 0.04455486658, 0.03786639547, 0.03747189194, 0.02769367774, 0.02256710184, -0.007460602423, -0.000360821725],
				[0.04501327517, 0.03441543433, 0.05846027012, 0.04801847343, 0.02887413717, 0.02797183226, -0.01440997349, -0.0003895354954],
				[0.04330779376, 0.03405688396, 0.04801847343, 0.04558680387, 0.03111517718, 0.02477230838, -0.01272882784, -0.0003624611328],
				[0.03019691712, 0.02516980916, 0.02887413717, 0.03111517718, 0.02614411029, 0.01475643353, -0.008794983792, -0.0002160623154],
				[0.02548665503, 0.02051044473, 0.02797183226, 0.02477230838, 0.01475643353, 0.01618991115, -0.006014483461, -0.0002507995642],
				[-0.01073017105, -0.006780679005, -0.01440997349, -0.01272882784, -0.008794983792, -0.006014483461, 0.005138124692, 0.00007878547574],
				[-0.0004006905689, -0.0003279381686, -0.0003895354954, -0.0003624611328, -0.0002160623154, -0.0002507995642, 0.00007878547574, 0.000007405024165]];
	var returns = [0.01807438883, 0.03238795043, 0.007555801824, 0.007427269126, 0.009034317809, 0.006707731718, 0.007769863126, 0.0007622417915];
	
	var minWeights = [0, 0, 0, 0, 0, 0, 0, 0];
	var maxWeights = [0.25, 0.25, 0.25, 0.25, 0.25, 0.25, 1, 1];

To Reproduce
C.f. above inputs.

Expected behavior
A solution to the mean variance optimization problem should be computed !

Link to Google Sheets / On-line npm code
N/A

Additional context
N/A

Hi. I am just wondering is there any real-world example? Don't really get how to use this package

Hello @lequant40, I'm trying to use this library for a personal project and it's proving a very valuable learning exercise for understanding the relevant concepts in Portfolio Optimization space. Thanks a lot for making this resource open source!

I was hoping to get some clarification on a bug I'm running into while using method meanVarianceEfficientFrontierPortfolios from lib/allocation/mean-variance.js. Here are the details -

Describe the bug
Getting the following error while using the method meanVarianceEfficientFrontierPortfolios -

ReferenceError: lambda_e is not defined
    at computeCornerPortfolios (portfolio_allocation.dist.js:15132:4)
    at new MeanVarianceEfficientFrontierCla (portfolio_allocation.dist.js:14596:26)
    at new MeanVarianceEfficientFrontierWrapper (portfolio_allocation.dist.js:16471:28)
    at self.meanVarianceEfficientFrontierPortfolios (portfolio_allocation.dist.js:17353:23)
    at getOptimalPortfolio (mvt-utils.ts?t=1699625436099:16:49)
    at App.tsx?t=1699626061247:25:17
    at commitHookEffectListMount (react-dom.development.js:23150:26)
    at commitPassiveMountOnFiber (react-dom.development.js:24926:13)
    at commitPassiveMountEffects_complete (react-dom.development.js:24891:9)
    at commitPassiveMountEffects_begin (react-dom.development.js:24878:7)

To Reproduce
I am trying to run this method on dummy data, like so -

let stockData = [[0.05, 0.01, 0.01], [0.1, -0.03, 0.05],]

let covMat = PortfolioAllocation.covarianceMatrix(stockData, {
  assumeZeroMean: false,
}).toRowArray();

let returns = [];

for (let s_i = 0; s_i < stockData.length; s_i += 1) {
  let R_i = PortfolioAllocation.returns(stockData[s_i]);
  let returns_i = 1;
  for (let s_i_t = 0; s_i_t < stockData[s_i].length - 1; s_i_t += 1) {
    returns_i *= 1 + R_i[s_i_t];
  }

  returns.push(returns_i - 1);
}

let mvEfficientFrontier =
  PortfolioAllocation.meanVarianceEfficientFrontierPortfolios(
    returns,
    covMat,
    { nbPortfolios: 10 }
  );

and got the aforementioned error.

Expected behavior
meanVarianceEfficientFrontierPortfolios method should return n portfolios (n=10 in this case) when called.

I would really appreciate it if you could look into this and provide some information. Thanks!

Hi Roman -

Thanks for this library! I'm just getting started with it on Google Sheets but can't seem to get the example in the README to work. In Google Sheets, I added your library with the Script ID, using static version 15 (v0.0.9). Then, I pasted this function into the Script Editor:

function computeERCPortfolioWeights(covarianceMatrix) {
// Note: The input range coming from the spreadsheet is directly usable.

// Compute the ERC portfolio weights
var ercWeights = PortfolioAllocation.equalRiskContributionWeights(covarianceMatrix);

// Return them to the spreadsheet
return (ercWeights);
}

In cells A1:B2 of my Google Sheet, I entered a sample covariance matrix: [[1,0.3],[0.3,1]]. In cell A4, I typed the following formula: =computeERCPortfolioWeights(A1:B2) but I get a #ERROR! with Error:unsupported input type (line 238).

I would've expected to get weights in cells A4:B4 or A4:A5.

Any help would be greatly appreciated.

Hello

Thanks a lot for this very cool library :)

I came across some NaN results while experimenting with the ERC algorithm. A Google Sheet with a specific test case can be found here:

https://docs.google.com/spreadsheets/d/1WNw4GLkoyOmC0Qjefga6N_imQejEIOtC_OjMXGWJkF0/edit?usp=sharing

I don't know whether this is supposed to happen, so please excuse this if it's a beginner's mistake.

Anyway, I thought I'd bring this to your attention.

Thanks again and best regards