TEJ Dictionary
Published in

TEJ Dictionary

【TEJ Dictionary】Does Short Selling Get High Returns?

Verify the predictability of Taiwan Short variables


As Bobby Axelrod, the main character in the drama “Billions”, said “ short-selling is just like leukocyte in the capital market. They specialize in swallowing bad companies, which can increase the efficiency of the capital market.” Since short-selling traders are usually “informed traders” who are capable of identifying overestimated stocks. Through short-selling, they can bring the stock price back to its fundamentals, which has the ability of price discovery.

In Taiwan, there are two kinds of short-selling ways, which are “SBL Short Sale” and “Margin Short Sale”. Most of the traders of “SBL Short Sale” are institutional investors, like foreign investors, investment trusts, and dealers. Unlike SBL Short Sale, traders of “Margin Short Sale” are mostly individuals and firms due to the regulations. Some literature indicates that “SBL Short Sale” and “Margin Short Sale” may have different influences on stock return due to the difference of traders.

Therefore, we discuss the expected return predictability of the FTSE TWSE Taiwan Mid-Cap 100 index’s constituent between two short-selling ways through the research process of factor-based investing, then choose the best factor of short selling.

🔸Highlights of this article🔸

✨Introduction of short selling in Taiwan✨

With collateral, Traders are able to borrow securities from an exchange or a broker-dealer. But having a deal of SBL does not mean short-selling. Therefore, we only provide some empirical evidence of the “SBL Short Sale” to reflect the true short-selling volume of investors. As for Margin Short Sale, traders can margin short sale through broker-dealer with enough margin.

✨Data & Variables✨

In this article, we take the constituents of FTSE TWSE Taiwan Mid-Cap 100 index as samples. The data were taken from TEJ API, and the period is from January 2013 to September 2021. Also, we treat the data properly to avoid survivorship bias and Look-ahead bias.

Referring to Lee et al.(2017), we organize short-selling variables into table 1 below. Considering the amount of “Margin Short Sale” and “SBL Short Sale” will affect the volume and market value of a stock, adjusting the volume and market value or taking growth rate as a short-selling variable is necessary. For detailed variable treatment, see table 1 below.

Table1: short-selling variables

We organize all the descriptive statistics into table 2 below. As you can see, all the short-selling variables are positive skewness, which represents that there are stocks short sold in high volume. We can also discover that “SBL Short Sale” takes more shares than “Margin Short Sale” in short selling. We presume the reason is that institutional investors short sell mainly through “SBL Short Sale”.

table 2: descriptive statistics of variables


To understand if the short-selling variables are effective factors, we take cross-sectional regression analysis and portfolio analysis as the research methodology.

In cross-sectional regression, we have a regression of the return rate of the next month(t) on the return rate of the current month(t+1). In table 3, among all the variables, we can see that only sbls_v has statistical significance, which means that there is a negative relation between sbls_v and the expected return of a stock, the others are statistically insignificant.

table 3: result of cross-sectional regression

As for portfolio analysis, we sort the variable by size and separate it into 5 equal-weighted portfolios, namely P1, P2, P3, P4, and P5. Next, we hold it for one month and calculate the average return rate, then change the stocks among the portfolios to see if it is monotonicity and has predictability between short selling variables and expected return. Figure 1 shows sbls_v has the best predictability, which is consistent with the result of cross-sectional regression analysis.

Figure 1: Chart of the monthly average return of P1 to P5


Through the analysis of cross-section regression, we can have a conclusion that sbls_v has a significant negative relation with the expected return, and sbls_v indeed has predictability to the expected return of the future, but others don’t.

In addition, the empirical result is inconsistent between the variables of “SBL Short Sale” and “Margin Short Sale”. The possible reason is that the trader of “lending securities” are mostly institutional investors who have preferable to the constituents of FTSE TWSE Taiwan Mid-Cap 100 index. While institutional investors are generally regarded as “informed traders”, the difference between traders of “SBL Short Sale” and “Margin Short Sale” could have different predictability of variables to expected return.

This article is summarized from “The relationships between short selling and stock return” written by Ricky

🔸Related link🔸

⭐️TEJ official website


⭐️ TEJ Instagram: tej87681088




Macroeconomics / Operation / Finance / Trade

Recommended from Medium

eDiscovery Data: What Else is it Good For?

[Datacamp] Intermediate Python — Dictionaries & Pandas

Electronic Health records — Multi-Document Summarization task

An interview riddle seen by a mathematician

Open Urban Mapping Update: Chechnya and Tyva

Data Viz Thoughts .|: Chris Conn Interview & Kate Schaub has our #VOTW

How to easily find geo data using Open Street Map

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store


TEJ為台灣本土第一大財金資訊公司,成立於 1990 年,專門提供金融市場基本分析所需資訊,以及信用風險、法遵科技、資產評價、量化分析及ESG等解決方案及顧問服務。鑒於財務金融領域日趨多元與複雜,TEJ 結合實務與學術界的精英人才,致力於開發機器學習ML、人工智慧AI及自然語言處理NLP等新技術,持續提供創新服務

More from Medium

Simple Investment Formulas

Thomas’ Proof of Value, January 2022

Is anyone else reading your Tinder messages in India?

Inversion and Trading