Table of Contents


File

probability_liquidation.m

Name

probability_liquidation

Synopsis

probability_liquidation - calculates the probability of fund liquidation. Takes as input coefficients generated from probability_liquidation_model.

Introduction

NOTE: PART OF A SET OF 2 RELATED FILES:

Chan, Getmansky, Haas, and Lo (2006a, 2006b) consider the broader impact of hedge funds on systemic risk by examining the unique risk/return profiles of hedge funds at both the individual-fund and aggregate-industry levels and propose three new risk measures for hedge fund investments. In the third risk measure, the authors create a measure of the probability of hedge fund liquidation by running a logit model on a set of factors driving hedge fund performance.

The authors describe three metrics in their paper; we concetrate on Hedge Fund Liquidation Probability. This metric is discussed extensively in Section F.7.1.2, and specifically in equations (A.106) and (A.107), in Bisias et al. (2012). Please refer to those pages for an extensive description of the code in this module and the required inputs.

License

=============================================================================

Copyright 2011, Dimitrios Bisias, Andrew W. Lo, and Stavros Valavanis

COPYRIGHT STATUS: This work was funded in whole or in part by the Office of Financial Research under U.S. Government contract TOSOFR-11-C-0001, and is, therefore, subject to the following license: The Government is granted for itself and others acting on its behalf a paid-up, nonexclusive, irrevocable, worldwide license to reproduce, prepare derivative works, distribute copies to the public, perform and display the work.
All other rights are reserved by the copyright owner.

THIS SOFTWARE IS PROVIDED "AS IS". YOU ARE USING THIS SOFTWARE AT YOUR OWN RISK. ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHORS, CONTRIBUTORS, OR THE UNITED STATES GOVERNMENT BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

=============================================================================

Inputs

coefficients
Name:
coefficients
Description:

Regression coefficients B0-B8 are inputted as a vector ordered as follows: B0 - Age; B1 - assets at time t-1; B2, B3, B4 - returns at time t, t-1, and t-2; B5, B6, B7 - flows at time t, t-1, and t-2; B8 - a y-intercept constant. Note: This is generated by probability_liquidation_model.m.

Type:
float
Range:
(-inf,+inf)
Dimensions:

9x1 matrix

  1. Rows represent different independent variables.

age
Name:
age
Description:

The age of the hedge fund.

Type:
float
Range:
(0,+inf)
Dimensions:

scalar


prev_assets
Name:
prev_assets
Description:

The assets under management of the hedge fund last year.

Type:
float
Range:
(0,+inf)
Dimensions:

scalar


returns
Name:
returns
Description:

The returns of the hedge fund in reverse chronological order; i.e. returns(1) are the current returns.
Note: This ordering is required to match the sequence of regression coefficients in Input arg coeff. While coeff has 9 values, its coefficients for returns are ordered in descending order as t, t-1, and t-2. Thus, returns which is multiplied times its respective coeff regression coefficients must also be descending.

Type:
float
Range:
(-inf,+inf)
Dimensions:

3x1 matrix

  1. Rows represent dates. (descending)

flows
Name:
flows
Description:

The flows of the last 3 years in reverse chronological order; i.e flows(1) are the current flows. Note: This ordering is required to match the sequence of regression coefficients in Input arg coeff. While coeff has 9 values, its coefficients for flows are ordered in descending order as t, t-1, and t-2. Thus, flows which is multiplied times its respective coeff regression coefficients must also be descending.

Type:
float
Range:
(-inf,+inf)
Dimensions:

3x1 matrix

  1. Rows represent dates. (descending)

Outputs

p
Name:
p
Description:

Estimated liquidation probability generated using regression equation and hedge-fund specific data.

Type:
float
Range:
(0,1)
Dimensions:

scalar


Code

% Run warning message
warning('OFRwp0001:UntestedCode', ...
    ['This version of the source code is very preliminary, ' ...
     'and has not been thoroughly tested. Users should not rely on ' ...
     'these calculations.']);


%
% Parameters:
% coefficient The 9 coeffs returned by the function 
% probability_liquidation_model
% age The age of the fund
% prev_assets The assets under management of the fund last year
% returns A 3x1 vector for the returns of the last 3 years going backwards
% i.e returns(1) are the current returns
% flows A 3x1 vector for the flows of the last 3 years going backwards
% i.e flows(1) are the current flows

linear_term = coefficients'*[1 age prev_assets returns' flows']';

p = exp(linear_term)/(1+exp(linear_term));

Examples

NOTE: Numbers used in the examples are arbitrary valid values.
They do not necessarily represent a realistic or plausible scenario.

 coefficients = [.11, -.07, .09, -.16, .08, .03, .05, -.08, .09]'
 age = 20
 prev_assets = 4.6
 returns = [0.05, 0.03, 0.1]'
 flows = [.9, .7, 1.3]'

 p = probability_liquidation(coefficients, age, prev_assets, ...
 returns, flows)

References

Chan et al. (2006a). Do hedge funds increase systemic risk?. Economic Review-Federal Reserve Bank of Atlanta, 91(4), 49.

Chan et al. (2006b). Systemic risk and hedge funds. The Risks of Financial Institutions. Chicago, IL: University of Chicago Press, 235-330.

Bisias et al. (2012). A survey of systemic risk analytics (Working paper #0001). Washington, DC: Office of Financial Research, 133-135.