17

Yes. First, it is much easier to proceed if you standardize the output of your forecast so they are in the same units (returns, for example, or probabilities of an event/condition occurring). After you have done this, there are 3 general approaches: Signal weighting: Then you need to define a weighting scheme for your factors. Richard Grinold has an one ...


12

Here are couple references. Especially the first link to Andy Lo's paper contains a list of Sharpe ratios of popular mutual and hedge funds: The Statistics of Sharpe Ratios Dow Jones Credit Suisse Hedge Fund Index Generalized Sharpe Ratios and Portfolio Performance Evaluation I would go with the first paper.


11

The answer your are looking for might be the story in "Benchmarking Measures of Investment Performance with Perfect-Foresight and Bankrupt Asset Allocation Strategies", by Grauer (Journal of Portfolio Management). While this work main concerns are the differential ranking of various performance measures and with negative betas for market timing strategies, ...


9

Perform a returns analysis by regressing the returns of your composite strategy on the returns of the component strategies. Constrain the beta coefficients to sum to 100% and bound them from 0 to 1. You will then have the % explained by each component.


7

We actually managed to come up with the answer to this question ourselves but wanted to share the answer since it might be relevant to others as well. The calculation depends on what method is used to calculate the cost. There is the FIFO, LIFO and the average cost method, see: http://www.accounting-basics-for-students.com/fifo-method.html If FIFO or LIFO ...


5

As you mention neural network, in general, you may like to look further into various machine learning techniques. On that side, Quant Guy also mentioned ensemble learning which is the general term to combine different learning models. I'd like to elaborate on this point a bit further: In machine learning, traditional ways to combine models are simple ...


5

Whatever method you use, I recommend you test your implementation with Monte Carlo simulations as well as real data (although doing the latter subjects you to data mining bias, it can give a sanity check on your Monte Carlo simulations.) For most instances of multiple algorithms, the returns streams will not be independent, and you should take this into ...


5

I would even stick to the original paper by Sharpe (1966): Mutual Fund Performance. The Journal of Business Vol. 39, No. 1, Part 2 pp.119--138 If you look at the numbers on Page 6 you can see that the funds sharpe ratios roughly are between $0$ and $1$. Since the Sharpe ratio already adjusts for the risk-free rate, you cannot really argue about its ...


5

For client reporting purposes, it is customary to use discrete returns. For backtesting, it pretty much make no difference.


4

Have you considered the HDF5 data model? Edit for Louis : Why using HDF5 ? As stated in the HFDF short description page : HDF5 is a unique technology suite that makes possible the management of extremely large and complex data collections. HDF5 is a suitable solution when dealing with very large datasets and you need performance. Again, as stated ...


3

Perhaps check out Poti and Levich (2009), or in a different setting but from one of the same authors, Poti and Wang (2010) "The coskewness puzzle" in JBF. They directly address the issue of what level of SR is plausible.


3

Pardon the lack of an actual link, and the formatting, but in footnote 6 of "Alpha is Volatility times IC times Score", Grinold, Richard C., Journal of Portfolio Management, Summer 1994 v20 n4 p9(8), Grinold suggests that "a truly outstanding manager" might have an information ratio of 1.33: (6) A rough guideline for determining the required IC comes from ...


3

This is a very common and serious problem among academic papers and with some hedge fund marketing material's, I can almost guarantee that the high ratio of 7 was with-out transaction cost's and that when included this 7 will drop down some where between 0 and 1.


2

You actually need to consider a 0 return on the periods with no holdings (during that period volatility is 0 and you have a negative return due to the opportunity cost of not holding risk free debt). From that you can compute your daily sharpe ratio and then multiply by $252^{0.5}$ as you mention.


2

If you do step 1 and step 2 every day, then you indeed assume that you rebalance the strategy every day. If you want to assume differently, for example monthly, you need to first compound the returns for each asset separately during the whole month and then do a weighted sum of the compounded returns using the weights of each asset at the beginning of the ...


2

Using Andy Flury answer and bit polishing it gives following Python class for PnL calculator: class PnLCalculator: def __init__(self): self.quantity = 0 self.cost = 0.0 self.market_value = 0.0 self.r_pnl = 0.0 self.average_price = 0.0 def fill(self, n_pos, exec_price): pos_change = n_pos - self....


2

I am not an expert on GIPS, with its many pages of rules, but I do remember that under GIPS Private Equity results are to be given in terms of IRR (Internal Rate of Return). In most other cases (stock/bond portfolios for example) GIPS requires TWR (Time Weighted Return) and forbids the use of IRR. To compute the IRR we need the dates and amounts of cash ...


2

The returns (or rather alphas, i.e. returns relative to the benchmark) plotted are logarithmic returns, not the simple returns usually reported by investment managers. This makes them additive over time. The green line is a straight line with slope 18% (the expected annual alpha). The thin purple curve is $\sigma \sqrt{t}$ above and below the green line, ...


2

There's nothing in the math that says a portfolio can only put non-zero weights on securities where the benchmark puts positive weights. So I'm not sure I understand your problem? Quick math review Let $R$ be a $k \times 1$ random vector denoting next period returns. Let $\mathbf{w}_b$ and $\mathbf{w}_b$ be $k \times 1$ vectors denoting weights of the ...


1

Sharpe is (Portfolio Return - RFR) / Standard Deviation. Information Ratio is (Portfolio Return - Benchmark Return) / Tracking Error, where tracking error is the standard deviation of the active return. I don't understand Professor X's comment either.


1

Holding period return would be more appropriate. Calculate your one week return by using your ending portfolio NAV. The easiest way to do this would to be to store number of shares in each position and multiply by price after one week to obtain your new NAV. Yes. Yes. No, subtract it from your ending portfolio NAV. The process is as follows: Estimate ...


1

It depends what you assume as to rebalancing between the portfolios. Unless these portfolios are being actively rebalanced each quarter to bring them back to exactly 32/68 allocation you should not assume fixed weights of 32 and 68. It is misleading. Here is an alternative calculation assuming no rebalancing: You start on 6/30/2018 with 0.32 million in ...


1

As a simple answer, the covariance matrix should not represent only assets in the benchmark. It should include the universe of assets. As an example, a benchmark might be 60% US Large Stocks and 40% US Aggregate Bonds. A manager might also buy emerging market stocks. One can just use a larger covariance matrix that includes emerging market stocks. The ...


1

These are the realized return and standard deviation for the portfolio over the period. Source: Paul Wilmott on Quantitative Finance, sec. ed., p. 329-330


1

Effective PA is dependent on the correct description of the investment process. I am not sure, from what you say, what exactly is your investment process. But let me presume that it is the following: You have chosen the S&P500 as your benchmark. You first distributed your money among sectors. (That you gave many sectors zero weight is not relevant to the ...


1

Maybe the following guidelines help: Big picture: For performance measurement purposes you should compare returns not absolute values. You need to convert all time series into percent returns which in itself takes care of normalization. Also as next step you do not only want to measure out or under performance in terms of return performance but in risk-...


1

Practically, the best metric is the one your boss wants you to use. Alternately, you can think of the return of the desk like a leveraged security, as described here. This would suggest that the daily performance would be calculated as the profit divided by the basis. If you want to express the return in terms of the capital requirement, then that's one ...


1

Can we make the assumption that the amount of each dividend is directly correlated to the amount of time between dividends? This would be the case if there were a fixed dividend payout ratio, or whatever the equivalent vocabulary is for this instrument. If so, you could annualize each dividend by the number of days since the previous dividend. The reason ...


1

If you are using the Bisons model the frequency of compounding should be as much as the the rebalance frequency of the portfolio. In your case to get meaningful results you will need to use daily compounding. One this to be careful is that you must take transaction effect into account when calculating the return. If you are using a software like FactSet it ...


1

Apache Cassandra would be a good fit for storing real-time intraday data. It's a partitioned row store, where rows are organized into table using a partition key. It you use a schema where you store data for one ticker per row with partitioning by day or month (it has a limit of 2B records in a row), the operations in your questions would be very performant....


Only top voted, non community-wiki answers of a minimum length are eligible