predict risk 副本
DESCRIPTION
predict riskTRANSCRIPT
![Page 1: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/1.jpg)
PREDICTING RISK FROM FINANCIAL
REPORTS WITH REGRESSION
In Association for Computational Linguistics 2009.
Human Language Technologies, pp.272–280
Authors: Kogan, S., Levin, D., Routledge, B.R., Sagi, J.S., and
Smith, N.A.
Presenter: Chen QingZhi
Date : 20120315
![Page 2: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/2.jpg)
OUTLINES
1.Introduction
2.Stock Return Volatility
3.Problem Formulation
4.Dataset
5.BaseLine and Evaluation Method
6.Experiments
7.Conclusion
![Page 3: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/3.jpg)
INTRODUCTION
![Page 4: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/4.jpg)
INTRODUCTION
4.Our motivation and task:1)In real world, people use financial report to predict the
financial risk of investment of that company by human
experience.
2)Given some financial reports ,we try to automatically
predict a continuous quantity known as stock return
volatility which is an measurement of financial risk
5.The output variable in this work is
uncontroversial and resources(both text and
volatility record) are easy to obtain.
![Page 5: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/5.jpg)
STOCK RETURN VOLATILITY
![Page 6: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/6.jpg)
STOCK RETURN VOLATILITY
3.Why volatility?We are trying to predict how stable its price will be over a
future time period , especially one year
4.Volatility is easier predicted than stock
performance and not subject to any kind of
human expertise and disagreement.
![Page 7: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/7.jpg)
PROBLEM FORMULATION
![Page 8: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/8.jpg)
PROBLEM FORMUALTION
3. SVR is a well-known method for training a
regression model
where C is a regularization constant and controls the training
error
![Page 9: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/9.jpg)
PROBLEM FORMULATION
In terms of kernel function:
Then use this formula to solve W
![Page 10: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/10.jpg)
DATASET
1.Form 10-K: mandated by US Securities
Exchange Commission
2.Subsection 7A: quantitative and qualitative
disclosures about market risk.So we filter other sections from the reports and keep the
most important part
3.For some reasons , not all of the documents
pass the filter at all: bankrupt , delist …
![Page 11: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/11.jpg)
DATASET
Table 1: Dimensions of the dataset used in this paper , after filtering and
tokenization.
The near doubling in average document size during 2002–3 is possibly due to the
passage of the Sarbanes-Oxley Act of 2002 in the wake of Enron’s accounting
scandal (and numerous others).
![Page 12: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/12.jpg)
DATASET
REPORT SAMPLE:
The following discussion and analysis of ABC’s consolidated financial condition and consolidated results of operation should be read in conjunction with ABC’s Consolidated Financial Statements and Notes thereto included elsewhere herein. This discussion contains certain forward-looking statements which involve risks and uncertainties . ABC’s actual results could differ materially from the results expressed in, or implied by, such statements. See “Regarding Forward-Looking Statements.”
![Page 13: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/13.jpg)
DATASET
![Page 14: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/14.jpg)
DATASET
Data preparation :
Tokenization was applied to the text, including
punctuation removal, down casing, collapsing
all digit sequences, and heuristic removal of
remnant markup
![Page 15: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/15.jpg)
BASELINES AND EVALUATION METHOD
![Page 16: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/16.jpg)
BASELINES AND EVALUATION METHOD
Measurement is mean squared error:
![Page 17: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/17.jpg)
EXPERIMENTS
![Page 18: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/18.jpg)
EXPERIMENTS
![Page 19: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/19.jpg)
EXPERIMENTS
Objective representation:
![Page 20: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/20.jpg)
EXPERIMENTS
![Page 21: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/21.jpg)
RESULTS
Table 2: MSE (Eq. 6) of different models on test data predictions. Lower values are better. Boldface denotes improvements over the baseline, and denotes significance compared to the baseline under a permutation test (p <0.05).
![Page 22: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/22.jpg)
EFFECTS OF SARBANES-OXLEY
Sarbanes-Oxley Act of 2002, which sought to
reform financial reporting, had a clear effect on
informativity.
![Page 23: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/23.jpg)
RECENT DATA IS MORE IMPORTANT
![Page 24: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/24.jpg)
INTERPRET WEIGHTS
Table 3: Most strongly-weighted terms in models learned from various time periods (LOG1P model with unigrams and bigrams). “#” denotes any digit sequence.
![Page 25: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/25.jpg)
EXPERIMENTS
![Page 26: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/26.jpg)
![Page 27: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/27.jpg)
EXPERIMENTS
For example :”estimates”, which averages one
occurrence per document even in the 1996–
2000 period, experiences the same term
frequency explosion, and goes through a
similar weight change, from strongly indicating
high volatility to strongly indicating low volatility
![Page 28: Predict risk 副本](https://reader034.vdocuments.site/reader034/viewer/2022052202/559c15391a28abac098b475a/html5/thumbnails/28.jpg)
CONCLUSION
1.Testbed in NLP.
2.How well improvement this paper is present?
3.Interpretable weight can how commendatory
or derogatory term have changed.
4.Can compare information released by
different texts.