How Information Sources Shift Stock Prices: Empirical Evidence from TCRI Watchdog

Apr 15 2026

Keywords：

alternative data Credit Risk signal TCRI Watch dog

Table of Contents

Preface

TCRI Watchdog (WD) converts complex news and announcements into standardized quantitative alternative data. Building on our research into “Official Announcements (Source P)” and “Media News (Source N),” we have confirmed that disclosure channels directly dictate the speed and structure of market digestion.

This chapter moves from macro “Event Categories” to micro “Source × Sub-category” dimensions to capture actionable Alpha within granular events. Focusing on high-sensitivity “Corporate Control Events,” we analyze the signal heterogeneity between Source P and Source N. We further demonstrate how these high-precision signals assist investors in optimizing entry timing and hedging strategies.

👉Struggling with unquantifiable market noise?
TCRI WD turns raw text into a structured quantitative database

Backtesting Framework

This study adopts a standard event study methodology to evaluate abnormal returns (AR) and cumulative abnormal returns (CAAR) surrounding event occurrences.

Research Scope and Event Design

Market coverage: Taiwan-listed and OTC companies (including delisted firms)
Data Source: TCRI Watchdog (WD), focusing on both announcement-based (P) and news-based (N) events
Sample period: January 1, 2018 to August 31, 2025
Research Scope (Source × Subcategory)：This chapter focuses on 2 representative event groups selected based on CAAR performance and statistical significance, covering both positive and negative signals:
- P-MO: Corporate governance announcements — control / agency issues
- N-MO: News — control / agency issues
Event grouping: Based on the WD event scoring system, events are grouped as follows:

Group	Event Score
negative	<0
-1	=-1
-2	=-2
-3	=-3
positive	>0
1	=1
2	=2
3	=3
neutral	=0

Event Study Design
- Estimation Window：t-270 ~ t-21
- Gap Days：10 days
- Event Window： t-10 to t+30

The post-event window is extended to +30 days to better capture signal persistence and tradability.

Event Categories and Backtesting Results

Compared to the previous chapter’s aggregate analysis at the category level, this section further decomposes events into specific subcategories to identify how different information sources and event types affect market reactions.

The analytical approach is adaptive:

When subcategories exhibit consistent return directions, results are aggregated to preserve both sample size and statistical robustness
When subcategories show heterogeneous behavior, the analysis focuses on representative and predictive subgroups to enhance precision

Based on this framework, 2 representative event groups are examined across governance, fundamentals, and capital structure dimensions.

P-MO: Announcement-Based Governance Events (Agency Issues)

P-MO events originate from material disclosures (P), including ownership changes, control transfers, and governance disputes.

These disclosures are released via the MOPS system, including both mandatory regulatory filings and exchange- or firm-initiated announcements, characterized by high standardization and rapid information transmission.

Based on this principle, this study focuses on four representative event combinations across three major dimensions—corporate governance, fundamentals, and capital structure—to analyze their individual impacts on stock price performance.

This group consists of seven subcategories:

Subcategory	Description
MO01	Insider / major shareholder share transfers
MO02	Waiver of subscription rights
MO03	Insufficient insider shareholding
MO04	External intervention / control disputes
MO05	Control transfer
MO06	Share pledging
MO07	Financial distress of insiders / major shareholders

Event Distribution

Figure 1 presents the distribution of event scores. A total of 849 events are observed:

Negative: 52.8%
Neutral: 42.9%
Positive: 4.4%

Figure 1：Event Sample Distribution (P-MO)

Table 2：Detailed Statistical Results (P-MO)

Group	Event Count	Mean AAR (%)	Final CAAR (%)
1	37	0.50	+19.05***
positive	37	0.50	+19.05***
neutral	364	0.12	+4.64***
negative	448	−0.06	−2.41**
−1	320	−0.10	−4.03***
−2	83	0.03	+1.27
−3	45	0.03	+1.16

CAAR Analysis

CAAR patterns exhibit clear directional separation, indicating strong discriminatory power of event intensity.

Positive events show significant cumulative gains
Negative events exhibit persistent downward trends

Figure 3：CAAR Group Comparison Chart (P-MO)

Effect Decomposition

Market reactions differ by group:

Positive and neutral events are partially anticipated (pre-event reaction)
Negative events are primarily post-event driven, indicating delayed pricing of governance risks

Table 4：Effect Distribution Table (P-MO)

Group	CAAR(t−1)%	CAAR(t=0)%	CAAR(t+30)%	AAR(t−1)%	AAR(t=0)%	Pre%	t=0%	Post%	Post-Ret%
1	11.52	13.86	+19.05	2.72	2.34	60.5	12.3	27.2	+5.19
positive	11.52	13.86	+19.05	2.72	2.34	60.5	12.3	27.2	+5.19
neutral	3.57	3.87	+4.64	1.02	0.30	76.9	6.5	16.6	+0.77
negative	−0.62	−0.51	−2.41	0.02	0.11	25.6	−4.5	78.9	−1.90
−1	−1.07	−1.04	−4.03	−0.13	0.02	26.4	−0.5	74.1	−2.99
−2	2.53	3.53	+1.27	0.48	1.00	199.6	78.7	−178.3	−2.26
−3	−3.50	−4.46	+1.16	0.22	−0.96	−301.6	−82.8	484.4	+5.62

Investment Implications

The positive group demonstrates strong return potential, with a total CAAR of +19.05% and post-event return of +5.19%, suggesting tradability even after disclosure. However, with 60.5% of returns realized pre-event, timing becomes critical.

The negative group shows a CAAR of −2.41%, with the −1 subgroup reaching −4.03% and post-event return of −2.99%, indicating clear post-event momentum. In contrast, the −2 and −3 groups lack sufficient statistical robustness.