hudson-and-thames/portfoliolab

PortfolioLab is a python library that enables traders to take advantage of the latest portfolio optimisation algorithms used by professionals in the industry.

portfoliolab's Issues

On Colab (interpolate keyerror)

KeyError                                  Traceback (most recent call last)
<ipython-input-19-f1e64dc1b990> in <module>()
      2 # Compute Passive Aggressive Mean Reversion with no given weights and epsilon of 0.3.
      3 pamr = PAMR(optimization_method=0, epsilon=0.3)
----> 4 pamr.allocate(asset_prices=us_equity, resample_by='W', verbose=True)
      5 
      6 # Compute Passive Aggressive Mean Reversion - 1 with given user weights and epsilon of 0.4.

5 frames
/usr/local/lib/python3.7/dist-packages/portfoliolab/online_portfolio_selection/base.py in allocate(self, asset_prices, weights, resample_by, verbose)
     80 
     81         # Initialize all variables.
---> 82         self._initialize(asset_prices, weights, resample_by)
     83 
     84         # Iterate through data and calculate weights.

/usr/local/lib/python3.7/dist-packages/portfoliolab/online_portfolio_selection/pamr.py in _initialize(self, asset_prices, weights, resample_by)
     52                                  'M' for Month. The inputs are based on pandas' resample method.
     53         """
---> 54         super(PAMR, self)._initialize(asset_prices, weights, resample_by)
     55 
     56         # Check that epsilon is greater than 0.

/usr/local/lib/python3.7/dist-packages/portfoliolab/online_portfolio_selection/base.py in _initialize(self, asset_prices, weights, resample_by)
    105         # Resample asset.
    106         if resample_by is not None:
--> 107             asset_prices = asset_prices.resample(resample_by).last()
    108 
    109         # Set asset names.

/usr/local/lib/python3.7/dist-packages/pandas/core/generic.py in resample(self, rule, axis, closed, label, convention, kind, loffset, base, on, level, origin, offset)
   8078         >>> df2 = pd.DataFrame(d2,
   8079         ...                    index=pd.MultiIndex.from_product([days,
-> 8080         ...                                                     ['morning',
   8081         ...                                                      'afternoon']]
   8082         ...                                                     ))

/usr/local/lib/python3.7/dist-packages/pandas/core/resample.py in <module>()
     34 
     35 
---> 36 class Resampler(_GroupBy, ShallowMixin):
     37     """
     38     Class for resampling datetimelike data, a groupby-like operation.

/usr/local/lib/python3.7/dist-packages/pandas/core/resample.py in Resampler()
    782         return self._upsample(method, limit=limit)
    783 
--> 784     @Appender(_shared_docs["interpolate"] % _shared_docs_kwargs)
    785     def interpolate(
    786         self,

KeyError: 'interpolate'

Cannot import Portfoliolab

Hi, I'm rather new in this library
I've pip insalled Portfoliolab as the documentation suggests but when tryng to import the module into a Jupyter notebook I receive the following errors:

import numpy as np
import pandas as pd
import portfoliolab as pl

untimeError                              Traceback (most recent call last)
RuntimeError: module compiled against API version 0xe but this version of numpy is 0xd

---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
<ipython-input-2-764cd2d99591> in <module>

...

**ImportError: numpy.core.multiarray failed to import**

I tried updating Numpy to version 1.20 but Portfoliolab requieres Numpy 1.18.5, so I really don't know what to do.

Can you help fix this?

Thank you and congratulations for these awesome libraries!

Data missing on "Online Portfolio Selection" notebook

On PortfolioLab - I follow the notebook example of Online Portfolio Selection part

msci = pd.read_csv('data/MSCI.csv', parse_dates=True, index_col='Date').dropna()[1:]
us_equity = pd.read_csv('data/US_Equity.csv', parse_dates=True, index_col='Date')
djia = pd.read_csv('data/DJIA.csv', parse_dates=True, index_col='Date')
nyse = pd.read_csv('data/NYSE.csv', parse_dates=True, index_col='Date')
sp500 = pd.read_csv('data/SP500.csv', parse_dates=True, index_col='Date')
tse = pd.read_csv('data/TSE.csv', parse_dates=True, index_col='Date')

Can you share the data which used in the notebook.
The provided link in the notebook is broken
https://github.com/hudson-and-thames/research/blob/master/Online%20Portfolio%20Selection/Data%20Selection.ipynb

DLASCL Parameter Error

ValueError                                Traceback (most recent call last)
<ipython-input-21-c0ce94737fd8> in <module>()
     20 # Compute Confidence Weighted Mean Reversion - Var with given user weights, epsilon of 1, and confidence of 1.
     21 cwmr_var = CWMR(confidence=1, epsilon=1, method='var')
---> 22 cwmr_var.allocate(asset_prices=us_equity)
     23 
     24 # Compute Online Moving Average Reversion - 1 with no given weights, epsilon of 10, and window of 7.

4 frames
<__array_function__ internals> in pinv(*args, **kwargs)

<__array_function__ internals> in svd(*args, **kwargs)

/usr/local/lib/python3.7/dist-packages/numpy/linalg/linalg.py in svd(a, full_matrices, compute_uv, hermitian)
   1659         The matrix whose condition number is sought.
   1660     p : {None, 1, -1, 2, -2, inf, -inf, 'fro'}, optional
-> 1661         Order of the norm:
   1662 
   1663         =====  ============================

ValueError: On entry to DLASCL parameter number 4 had an illegal value

Cannot pip install portfoliolab on google colab

• portfoliolab
• Python version: 3.7
• Operating System: Windows

I tried '!pip install portfoliolab' using google colabs but it returns:

"ERROR: Could not find a version that satisfies the requirement portfoliolab (from versions: none)
ERROR: No matching distribution found for portfoliolab"

It was working for me about 1 week ago, but maybe some update changes have been made which result in this error.