How News and Its Context Drive Risk and Returns

How News and Its Context Drive Risk and Returns

Around the World

Charles W. Calomiris and Harry Mamaysky

Columbia Business School

1

Introduction

• Automated processing of natural language is opening a previously unavailable window into market behavior

• It may fundamentally transform finance practice

• Prior work has been very short-term focused

• But isn’t news (in aggregate) important for longer horizon outcomes?

• We look at

• Longer term country-level risk and return responses to news

• How to measure news at the country level?

2

Our approach and a peak at findings…• We develop a theory-neutral approach to map country news into market

outcomes, which measures word flow and examines connections of word flow to risk and return.

• We apply this (for the first time, we think) outside the U.S., to 52 countries.

• EMs vs. DMs treated separately, given differences in returns processes.

Key Findings:

1. Many measures relevant (sentiment, frequency, entropy), EMs/DMs differ.

2. Topical context matters.

3. Results change over time importantly.

4. News generally has opposite implications for return and risk.

5. Drawdown is useful as a measure of risk, especially for EMs.

6. We capture more than a popular a priori measure, in and out of sample.

3

1. Theory-neutral vs. a priori word identifiers

What word flow?

• Theory-neutral vs. a priori approaches (Baker Bloom Davis 2016)

• Theory-neutral does not require advance knowledge of what is important, and avoids data mining risks.

• But is it possible to construct a comprehensive, parsimonious, and flexible theory-neutral model of word flow?

4

2. What aspects of news are important?

• Sentiment

• Frequency

• Unusualness (entropy)

• Interact sentiment and entropy (Glasserman and Mamaysky 2016)

• Topical context interacted with above• How are topics different from EM to DM?

• How does effect of news, and interpretation of news, differ by topic?

5

3. Regime changes over time?

• Principal components indicate shift point around Global Crisis

• A priori shift point lines up with second principal component

• Out of sample properties of forecasting in light of this change

6

4. How to identify topical context?

• Identifying topic-relevant words and their characteristics

• Louvain method vs. LDA

• Computational difficulty of LDA (pilot study comparison)

7

5. Is all news relevant for both returns and risk?

• Allow our measures to affect both returns and risk and see whether effects tend to be opposite, or unrelated.

• Will we find opposite signs when an effect is statistically significant for return, if it is also statistically significant for sigmaor drawdown?

8

6. How to measure risk?

• Especially in EMs, returns are not normal and there is momentum in returns.

• In addition to sigma, we use drawdown (which allows longer term effects from momentum, skew, and kurtosis to be expressed).

9

7. How to analyze countries, together or not?

• Advantages to panel.

• Disadvantages from pooling (if processes are different)

• We separate EMs and DMs and analyze each as a panel.

10

8. What news source?

• Thomson-Reuters provides a common platform, English language, and large sample of relevant countries, for which there are other data on returns and on various relevant variables.

11

Data construction

• Data cleaning

• Financial economics word list (Beim-Calomiris plus)

• Topics (Louvain groups), EM vs. DM topic groups

• Sentiment (Loughran-McDonald)

• Entropy (4-grams)

• Context-specific measures of sentiment, frequency, entropy

12

Text measures definedData

• Thomson-Reuters digital news archive from 1996—2015

• 5mm EM and 12mm DM articles

• 52 countries (list next page)

Text measures:

• artcount – number of articles per country per month

• entropy – “unusualness” of an article j (Glasserman and Mamaysky 2016)

𝐻𝑗 = −

𝑖 ∈ {4−grams}

𝑝𝑖 log𝑚𝑖

• Effectively the average log probability of a word conditional on preceding words

• sentiment – the difference of positive and negative words divided by total words in article j:

𝑠𝑗 =𝑃𝑂𝑆𝑗 − 𝑁𝐸𝐺𝑗

𝑎𝑗

• Word sentiment comes from Loughran – McDonald dictionary13

Topics

Intuition: Find groups of words that co-occur together in articles

Details:• 1240 econ words

• Start w/ 237 words from index of Beim and Calomiris (2001) and find other words, bigrams and trigrams from EM corpus based on cosine similarity

• E.g.: barriers, currency, parliament, macroeconomist, and World Bank

• We have 2 document term matrixes:

• 5mm x 1,240 for EM and 12mm x 1,240 for DM

• Compute cosine similarity matrix (1,240 x 1,240)

• Then do community detection (using Louvain method for modularity maximization)

• Out topics are mutually exclusive (not necessary)

14

We find 5 topics for each group of countries

• The Louvain algorithm returns ~40 word clusters with the following numbers of words

• Place words from small clusters into big clusters

15

Topics for EMs

16

Topics for DMs

17

Context specific sentiment

• Let 𝑓𝜏,𝑗 be the fraction of econ words in article j that are about topic τ

• Topic sentiment is given by: 𝑠𝜏,𝑗 = 𝑓𝜏,𝑗 × 𝑠𝑗

• Aggregate the article level measures into daily measures (weighted by number of overall words in an article)

For a given country, we have 12 daily text measures:• entropy• article count• sMkt / fMkt• sGovt / fGovt• sCorp / fCorp• sComms / fComms• DM/EM specific:

• sMacro / fMacro (EM)• sCredit / fCredit (DM)

18

Principal Components EM

19

EM Sentiment

• For 140 EM sentiment series (28 countries x 5) we look at first 2 principal components

• PC2 – relative sentiment of Markets to Government

• Some evidence of a regime shift in PC2 a little before the financial crisis

Principal components EM

20

DM Sentiment

• For 120 DM sentiment series (24 countries x 5) we look at first 2 principal components

• PC2 – relative sentiment of Markets to Government (again!)

• Some evidence of a regime shift in PC2 a little before the financial crisis

Event Studies

• High-frequency top and bottom deciles of sentiment

• Middle as placebo

• Returns lead major sentiment indicators at high frequency

• Some post-event drift for positive and negative events

21

Event studies – EM

• Cumulative abnormal return around deciles of daily news events

• Middle column is control for boring news

• Some topics show post event drift: Mkt (both), Comms (negative)

• This is very differentfrom single name results, where there is little evidence of drift post negative news (only post positive)!

22

Event studies – DM

• Some topics show post event drift: Mkt(negative, both?), Corp(positive), Credit (both)

23

Regression results

• We run panel regression with dependent variables given by

• return

• return12

• sigma

• drawdown

• We control for many variables that have been shown to have some forecasting power for future returns (next page)

• The no-text measure regression is our Baseline model

• All text measures (except entropy) are normalized to unit variance

• We run full sample, 1st and 2nd half of the sample

24

Control variables

25

Summary of regression results

• News matters for EM and DM!

• Results differ across EM and DM (e.g., artcount matters in EMs)• Baseline R2 lower for EM

• % increase in R2 from text measures larger for EM

• Sign of effects (i.e. good news or bad) almost always is consistent across return, sigma, and drawdown

• Context matters: positive sentiment in Govt, Corp – bad news; positive sentiment in Mkt – good news

• Incremental explanatory power largest for return12 and drawdown; explanatory power lower for return and sigma

• Evidence of state dependence, especially for entropy• Goes from a “bad” pre-crisis to a “good” post-crisis

26

Summary of regression results

27

28

29

Out-of-sample testing

• Do we have too many explanatory variables?

• What about regime shifts?

• Check out-of-sample forecasting performance

• Run rolling 5-year regressions in t-60,…,t-1 for forecasting month toutcomes

Lasso (least absolute shrinkage and selection operator)

min𝛽

1

2𝑁

1,𝑡

𝑦𝑖,𝑡 − 𝑥𝑖,𝑡−1′ 𝛽

2+ 𝜆 𝛽 1

• Lasso does shrinkage and model selection• Amount of shrinkage given by 𝛽 1/ 𝛽

𝑂𝐿𝑆1

30

When the model has little to say

31

Rolling lasso for DM drawdown

32

Rolling lasso for EM drawdown

33

Out-of-sample performance

• Naïve model forecasts using country fixed effects

• Base model includes only the non-text variables

• CM includes all text measures

• All models estimated using lasso

34

Out-of-sample comparisons to EPU

• EPU counts articles from 10 major papers that contain triplets from

uncertainty x

economic x

{policy terms}

• For 5 EM and 11 DM countries where we have EPU data, compare out-of-sample performance of Base vs Base + alternative text measures

35

Conclusions

• Useful information in text for medium-term country-level outcomes

• Different dimensions of text matter• In particular, context

• Effects differ across EM and DM, and over time

• Evidence of out-of-sample forecasting ability

• Next:• Currency effects?• Trading strategies?

36