01 · VARIABLE CONTRACT
연구 변수 정의 및 실증분석 설계
분석단위: 연도-기업 단위 10-K report
1. Dependent variables
| Variable | Symbol | Measurement | Source |
|---|---|---|---|
| Abnormal Return | AR | ARᵢτ = DlyRetᵢτ − (α̂ᵢ + β̂ᵢ VWRETDτ) | Pan et al. (2018); Ertekin, Sorescu, & Houston (2018) |
| Cumulative Abnormal Return | CAR | CARᵢ,[a,b] = Σ ARᵢτ | Pan et al. (2018); Ertekin, Sorescu, & Houston (2018) |
| Buy-and-Hold Abnormal Return | BHAR | Π(1+Rᵢτ) − Π(1+RBenchmark,ᵢτ) | Ertekin, Sorescu, & Houston (2018) |
현재 V8의 primary market DVs는 AR 및 CAR이며, BHAR은 후속 분석 후보이다. CAR windows: [-1,+1], [-2,+2], [-3,+3], [-1,+2], [-2,0], [0,+2].
2. Focal independent variables — AI communication
| Variable | Symbol | Measurement | Source |
|---|---|---|---|
| AI Mention | AI_Mention | 10-K 또는 해당 section에 AI-related term이 하나 이상 존재하면 1, 아니면 0 | Mishra, Ewing, & Cooper (2022) |
| AI Focus | AI_Focus | (No. of AI-related words / Total no. of words) × 100 | Mishra, Ewing, & Cooper (2022) |
| AI disclosure start | StartAI | 연속 관측연도에서 AI_Mention: 0→1. Prior-year missing은 0으로 대체하지 않음 | 본 연구에서 구성 |
| AI disclosure stop | StopAI | 연속 관측연도에서 AI_Mention: 1→0. Reference는 NoChange(0→0, 1→1) | 본 연구에서 구성 |
AI_Mention과 AI_Focus는 실제 AI adoption이 아니라 text-based AI disclosure / communication proxy로 해석한다.
3. Focal independent variables — Concreteness
| Variable | Symbol | Measurement | Source |
|---|---|---|---|
| Overall Concreteness | Concreteness_All | 전체 10-K의 Brysbaert dictionary-matched words에 대해 ΣCw / Nmatched | Brysbaert, Warriner, & Kuperman (2014); Baek, Ihm, & Kang (2023) |
| AI-sentence Concreteness | Concreteness_AI | AI-related term이 하나 이상 포함된 모든 문장의 dictionary-matched word score 평균 | Brysbaert et al. (2014); Baek, Ihm, & Kang (2023) |
| Non-AI Concreteness | Concreteness_NonAI | AI-related term이 없는 문장의 dictionary-matched word score 평균 | Brysbaert et al. (2014); Baek, Ihm, & Kang (2023) |
| Concreteness Difference | Delta_Concreteness | Concreteness_AI − Concreteness_NonAI | Baek, Ihm, & Kang (2023) 기반 본 연구 구성 |
Concreteness_AI는 AI keyword 자체의 점수가 아니라 AI-related sentences 전체의 matched-word average다. AI_Mention=0이면 AI_Focus=0이지만 Concreteness_AI와 Delta_Concreteness는 구조적으로 NA이며 0으로 대체하지 않는다.
4. Focal independent variables — TENSE
| Scope | Variables | Measurement / construction | Source |
|---|---|---|---|
| Overall | PastFocus, PresentFocus, FutureFocus, TimeFocusing | 각 focus = 해당 marker / Total words ×100; TimeFocusing = PastFocus − (PresentFocus + FutureFocus) | Pan et al. (2018); Baek & Ihm (2021) |
| AI sentences | PastFocus_AI, PresentFocus_AI, FutureFocus_AI, TimeFocusing_AI | 122-term AI-related sentences에 동일 계산 적용 | Pan et al. (2018); Baek & Ihm (2021); Mishra, Ewing, & Cooper (2022) |
| Non-AI sentences | PastFocus_NonAI, PresentFocus_NonAI, FutureFocus_NonAI, TimeFocusing_NonAI | AI term이 없는 문장에 동일 계산 적용 | Pan et al. (2018); Baek & Ihm (2021); Mishra, Ewing, & Cooper (2022) |
| AI − Non-AI | Delta_PastFocus, Delta_PresentFocus, Delta_FutureFocus, Delta_TimeFocusing | 각 AI value − NonAI value | Pan et al. (2018); Baek & Ihm (2021) 기반 본 연구 구성 |
현재 구현: PastFocus = VBD, PresentFocus = VBP/VBZ, FutureFocus = AUX will/shall/'ll/’ll. AI/Non-AI partition은 Concreteness_AI와 동일한 122-term boundary-aware matcher를 사용한다. AI_Mention=0일 때 AI-sentence TENSE 및 Delta_*Focus는 회귀에서 missing으로 처리하며 0으로 해석하지 않는다. 같은 scope의 Past/Present/Future와 TimeFocusing은 정확한 선형결합 관계이므로 동일 specification에 기계적으로 동시 투입하지 않는다.
5. Baseline textual controls
| Variable | Symbol | Measurement | Source |
|---|---|---|---|
| Fog Index | FOG | 0.4 × (Average words per sentence + Percentage of complex words) | Loughran & McDonald (2014); Ertugrul et al. (2017) |
| 10-K File Size | FileSize | SEC 10-K filing file size; 현재 회귀 실행에서는 ln(FileSize) 사용 | Loughran & McDonald (2014); Ertugrul et al. (2017) |
| Word Count | WordCount | Total words in 10-K; 현재 회귀 실행에서는 ln(TotalWords) 사용 | Baek, Ihm, & Kang (2023) |
| Positive Tone | PositiveTone | LM positive words / total words ×100 | Loughran & McDonald (2011); Pan et al. (2018) |
| Negative Tone | NegativeTone | LM negative words / total words ×100 | Loughran & McDonald (2011); Pan et al. (2018) |
6. Baseline firm-level controls
| Variable | Symbol | Measurement | Source |
|---|---|---|---|
| Firm Size | SIZE | ln(Total Assets) | Mushtaq et al. (2022) |
| Cash Holdings | CASH | Cash & Equivalents / Total Assets | Mushtaq et al. (2022) |
| Leverage | LEV | Total Debt / Total Assets | Mushtaq et al. (2022) |
현재 실제 회귀 사양에서 사용된 firm-level controls는 SIZE, CASH, LEV이다.
7. Candidate controls
| Variable | Symbol | Measurement | Source |
|---|---|---|---|
| Liquidity | LIQ | Current Assets / Current Liabilities | Mushtaq et al. (2022) |
| Financing Needs / Deficit | DEF | Dividend + Capital Expenditure + Change in Net Working Capital + Short-Term Debt + Current Portion of Long-Term Debt − Net Cash Flow | Mushtaq et al. (2022) |
| Research & Development | RD | R&D Expenses / Total Assets | Mushtaq et al. (2022) |
| Tangibility | TANG | Net PPE / Total Assets | Mushtaq et al. (2022) |
| Return on Assets | ROA | Income before extraordinary items / lagged total assets | Ertugrul et al. (2017); Pan et al. (2018) |
| Profitability | PROFITABILITY | EBITDA / Total Assets | Ertugrul et al. (2017) |
| Market-to-Book | MB | Market value of equity / Book value of equity | Ertugrul et al. (2017) |
| Turnover Change | DTURN | Change in average monthly share turnover | Ertugrul et al. (2017) |
| Return Volatility | SIGMA | Standard deviation of firm-specific weekly returns | Ertugrul et al. (2017) |
| Prior Return | RET | Mean firm-specific weekly returns | Ertugrul et al. (2017) |
| Expected Default Frequency | EDF | Expected Default Frequency | Ertugrul et al. (2017) |
| Firm Age | FirmAge | ln(years since first CRSP appearance) | Ertugrul et al. (2017) |
| Business Segment Index | BSEG | Sum of squared business-segment proportions | Ertugrul et al. (2017) |
| Reporting Opacity | OPAQUE | Absolute discretionary accruals using modified Jones model | Ertugrul et al. (2017) |
8. Fixed Effects and additional selection check
| Variable / effect | Symbol | Measurement | Source |
|---|---|---|---|
| Firm Fixed Effects | αᵢ | Firm-specific fixed effect | 본 연구의 baseline specification |
| Year Fixed Effects | δₜ | Year dummies | 본 연구의 baseline specification |
| Inverse Mills Ratio | IMR_hat | Probit first stage의 φ(Ẑᵢₜ)/Φ(Ẑᵢₜ) | Ertekin, Sorescu, & Houston (2018); Moon, Tuli, & Mukherjee (2023) |
02 · MODEL SPECIFICATION
기본 회귀식
모든 핵심 모형은 text controls, event-aligned t−1 firm controls, Firm FE(αᵢ), Year FE(δₜ)를 포함한다. 추정은 statsmodels.api.OLS(...).fit(), 기본 covariance는 nonrobust이다.
전체 AI 존재
CARᵢₜ = β₀ + β₁ AI_Mentionᵢₜ + γ′Tᵢₜ + θ′Fᵢ,ₜ₋₁ + αᵢ + δₜ + εᵢₜ
Section AI intensity
CARᵢₜ = β₀ + β₁ Section_AI_Focusᵢₛₜ + γ′Tᵢₜ + θ′Fᵢ,ₜ₋₁ + αᵢ + δₜ + εᵢₜ
변화 / 최초 언급
CARᵢₜ = β₀ + β₁ ΔAI_Mentionᵢₜ + controls + FE + εᵢₜ
CARᵢₜ = β₀ + β₁ First_AI_Mentionᵢₛₜ + controls + FE + εᵢₜ
Start / Stop decomposition
CARᵢₜ = β₀ + βS StartAIᵢₜ + βT StopAIᵢₜ + γ′Tᵢₜ + θ′Fᵢ,ₜ₋₁ + αᵢ + δₜ + εᵢₜ
Start / Stop × AI_Focus
CARᵢₜ = β₀ + β₁StartAI + β₂StopAI + β₃AI_Focusₜ + β₄AI_Focusₜ₋₁ + β₅(StartAI×AI_Focusₜ) + β₆(StopAI×AI_Focusₜ₋₁) + controls + FE + ε
Section transition × language
Start: CAR = β₀ + β₁StartAIᵢₛₜ + β₂Mᵢₛₜ + β₃(StartAI×M) + controls + FE + ε
Stop: CAR = β₀ + β₁StopAIᵢₛₜ + β₂Mᵢₛ,ₜ₋₁ + β₃(StopAI×Mₜ₋₁) + controls + FE + ε
M은 Concreteness_AI, FutureFocus_AI, PastFocus_AI를 각각 별도 추정. Stop은 1→0과 Stay1(1→1)을 비교하고 moderator는 t−1 사용.
03 · DESCRIPTIVE STATISTICS
표본 및 기술통계
Raw master
11,050
11,050
2026 filing 제외 후
10,656
10,656
Regression spine
10,135
10,135
Transition eligible pairs
9,082
9,082
Start
521
521
Stop
288
288
| Variable | N | Mean | SD | Median |
|---|---|---|---|---|
| AI_Mention | 10,135 | 0.6111 | 0.4875 | 1 |
| AI_Focus (%) | 10,135 | 0.00716 | 0.01661 | 0.00205 |
| Concreteness_All | 10,135 | 2.9486 | 0.0608 | 2.9430 |
| Concreteness_AI | 6,192 | 3.0069 | 0.2869 | 2.9946 |
| PastFocus_AI (%) | 6,193 | 0.7242 | 1.3144 | 0 |
| FutureFocus_AI (%) | 6,193 | 0.1266 | 0.4808 | 0 |
| FOG | 10,135 | 22.4208 | 1.1184 | 22.4322 |
| LEV | 9,531 | 0.2857 | 0.2008 | 0.2658 |
| SIZE | 9,559 | 9.8321 | 1.3702 | 9.7118 |
| CASH | 9,559 | 0.1247 | 0.1326 | 0.0768 |
표본 해석. 10,135는 raw HTML 수가 아니라 2026 filing 제외 및 duplicate audit 후 회귀 spine이다. AI_Mention 평균 0.611은 이 표본의 약 61.1% firm-year에서 AI가 한 번 이상 언급되었음을 뜻한다.
04 · CORRELATION & MULTICOLLINEARITY
상관계수 및 VIF
StartAI / StopAI specification complete-case Pearson correlation N=8,966. 가장 큰 절대 상관은 ln(FileSize)–ln(TotalWords) r=0.7109이다. StartAI–StopAI는 r=-0.0447이다.
| Regressor | VIF range |
|---|---|
| ln(TotalWords) | 2.294–2.297 |
| ln(FileSize) | 2.165 |
| SIZE | 1.218 |
| FOG | 1.190 |
| CASH | 1.176 |
| StartAI | 1.005 |
| StopAI | 1.004 |
Interaction specification 주의. Concreteness interaction에서 Event dummy와 interaction의 VIF가 높다(Start 약 105.6–131.9; Stop 약 77.1–119.8). 이는 비중심화된 interaction 구조와 관련된다. 변수 삭제나 mean-centering은 임의로 수행하지 않았다. Future/Past interaction VIF는 대체로 약 1–2 수준이다.
05 · MODEL-FREE EVIDENCE
Model-Free Evidence
회귀모형을 적용하기 전의 기술적·시각적 증거를 정리한다. 아래에는 filing event 전후 raw price path, AI communication의 연도별 분포, Concreteness의 연도별 변화, TENSE/temporal-focus의 연도별 변화가 포함된다. 2026 filing은 제외하고, -9999 sentinel은 missing으로 처리하며 임의 보간이나 0 대체를 하지 않는다.
Figure 3 · Model-Free Evidence Around V8 Filing Event, Trading Day -30 to +30
Ready events
10,020
10,020
Valid t0 price
9,976
9,976
Complete 61-day
9,917
9,917
t=0은 V8_EventTradingDateV8이고 종축은 log(Pₜ)-log(P₀)이다. Pre daily overall mean 평균은 -0.0010603, Post 평균은 -0.0025580이며 Post−Pre는 -0.0014977 log point(단순 환산 약 -0.150%)이다.
Colab STEP V8-5P 원본 Figure. CAR/AR가 아니며 AI 집단별 회귀효과를 나타내는 그림이 아니다.
Figure 4 · Model-Free Evidence Around V8 Filing Event, Trading Day -7 to +7
Ready events
10,020
10,020
Complete 15-day
9,962
9,962
Post−Pre
-0.004534
-0.004534
Pre 평균은 0.0018613, Post 평균은 -0.0026723이며 단순 환산 약 -0.452%이다.
Colab STEP V8-5Q 원본 Figure. 이 -0.452%는 StartAI 회귀계수가 아니다.
Figure 5 · AI_Mention by Year
Colab STEP V8-5B. 각 연도에서 AI_Mention=0과 AI_Mention=1의 관측치 수를 별도로 표시한다.
Figure 6 · Mean AI_Focus by Year
Colab STEP V8-5B. AI_Focus의 연도별 평균을 표시한다.
Figure 7 · Annual Mean Concreteness — All vs AI vs NonAI
Colab STEP V8-5D. Concreteness_All을 reference로 두고 AI 및 Non-AI sentence concreteness의 절대 수준을 연도별로 비교한다.
Figure 8 · Annual Mean Concreteness Gaps vs All
Colab STEP V8-5D. 동일 raw row에서 두 값이 모두 존재하는 paired comparison을 사용하여 AI−All 및 NonAI−All gap을 표시한다.
Figure 9 · Annual Mean Delta_Concreteness (AI − NonAI)
Colab STEP V8-5D. Delta_Concreteness = Concreteness_AI − Concreteness_NonAI의 연도별 평균을 별도 축에서 표시한다.
Figure 10 · Annual Mean Tense / Temporal Focus — WHOLE
Colab STEP V8-5E. 전체 10-K scope의 PastFocus, PresentFocus, FutureFocus, TimeFocusing 연도별 평균.
Figure 11 · Annual Mean Tense / Temporal Focus — AI
Colab STEP V8-5E. AI-related sentence scope의 TENSE / temporal-focus 연도별 평균.
Figure 12 · Annual Mean Tense / Temporal Focus — NONAI
Colab STEP V8-5E. Non-AI sentence scope의 TENSE / temporal-focus 연도별 평균.
Figure 13 · Annual Mean Tense / Temporal Focus — DELTA
Colab STEP V8-5E. AI − NonAI TENSE / temporal-focus 차이의 연도별 평균이며, 0 기준선을 함께 확인한다.
06 · OLS TABLE
OLS Table
독립변수, 종속변수, 오차항
CAR[-1,+1] | N=9,910
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m1_p1 R-squared: 0.000
Model: OLS Adj. R-squared: 0.000
Method: Least Squares F-statistic: 1.102
Date: Tue, 22 Sep 2026 Prob (F-statistic): 0.347
Time: 04:31:54 Log-Likelihood: 18048.
No. Observations: 9910 AIC: -3.609e+04
Df Residuals: 9906 BIC: -3.606e+04
Df Model: 3
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept 0.0008 0.002 0.448 0.654 -0.003 0.005
AI_Mention 0.0007 0.001 0.795 0.427 -0.001 0.002
Concreteness_All__MM01 0.0025 0.004 0.636 0.525 -0.005 0.010
TimeFocusing__MM01 -0.0056 0.004 -1.514 0.130 -0.013 0.002
==========================================================================================
CAR[-2,+2] | N=9,905
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m2_p2 R-squared: 0.000
Model: OLS Adj. R-squared: 0.000
Method: Least Squares F-statistic: 1.443
Date: Tue, 22 Sep 2026 Prob (F-statistic): 0.228
Time: 04:31:54 Log-Likelihood: 16179.
No. Observations: 9905 AIC: -3.235e+04
Df Residuals: 9901 BIC: -3.232e+04
Df Model: 3
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept 0.0007 0.002 0.317 0.751 -0.004 0.005
AI_Mention 0.0011 0.001 1.111 0.267 -0.001 0.003
Concreteness_All__MM01 0.0048 0.005 1.012 0.312 -0.005 0.014
TimeFocusing__MM01 -0.0070 0.004 -1.574 0.116 -0.016 0.002
==========================================================================================
CAR[-3,+3] | N=9,902
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m3_p3 R-squared: 0.000
Model: OLS Adj. R-squared: 0.000
Method: Least Squares F-statistic: 1.405
Date: Tue, 22 Sep 2026 Prob (F-statistic): 0.239
Time: 04:31:54 Log-Likelihood: 14633.
No. Observations: 9902 AIC: -2.926e+04
Df Residuals: 9898 BIC: -2.923e+04
Df Model: 3
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept 0.0006 0.003 0.243 0.808 -0.005 0.006
AI_Mention 0.0016 0.001 1.339 0.181 -0.001 0.004
Concreteness_All__MM01 0.0048 0.006 0.869 0.385 -0.006 0.016
TimeFocusing__MM01 -0.0071 0.005 -1.357 0.175 -0.017 0.003
==========================================================================================
CAR[-1,+2] | N=9,905
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m1_p2 R-squared: 0.000
Model: OLS Adj. R-squared: 0.000
Method: Least Squares F-statistic: 1.241
Date: Tue, 22 Sep 2026 Prob (F-statistic): 0.293
Time: 04:31:54 Log-Likelihood: 17009.
No. Observations: 9905 AIC: -3.401e+04
Df Residuals: 9901 BIC: -3.398e+04
Df Model: 3
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept 0.0010 0.002 0.479 0.632 -0.003 0.005
AI_Mention 0.0004 0.001 0.466 0.641 -0.001 0.002
Concreteness_All__MM01 0.0043 0.004 0.982 0.326 -0.004 0.013
TimeFocusing__MM01 -0.0072 0.004 -1.758 0.079 -0.015 0.001
==========================================================================================
CAR[-2,0] | N=9,911
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m2_0 R-squared: 0.000
Model: OLS Adj. R-squared: -0.000
Method: Least Squares F-statistic: 0.5803
Date: Tue, 22 Sep 2026 Prob (F-statistic): 0.628
Time: 04:31:54 Log-Likelihood: 18329.
No. Observations: 9911 AIC: -3.665e+04
Df Residuals: 9907 BIC: -3.662e+04
Df Model: 3
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept 0.0005 0.002 0.290 0.772 -0.003 0.004
AI_Mention 0.0001 0.001 0.172 0.864 -0.001 0.002
Concreteness_All__MM01 0.0040 0.004 1.049 0.294 -0.004 0.012
TimeFocusing__MM01 -0.0037 0.004 -1.042 0.298 -0.011 0.003
==========================================================================================
CAR[0,+2] | N=9,905
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_0_p2 R-squared: 0.001
Model: OLS Adj. R-squared: 0.000
Method: Least Squares F-statistic: 1.682
Date: Tue, 22 Sep 2026 Prob (F-statistic): 0.169
Time: 04:31:55 Log-Likelihood: 18880.
No. Observations: 9905 AIC: -3.775e+04
Df Residuals: 9901 BIC: -3.772e+04
Df Model: 3
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept 0.0005 0.002 0.307 0.759 -0.003 0.004
AI_Mention 0.0012 0.001 1.504 0.133 -0.000 0.003
Concreteness_All__MM01 0.0012 0.004 0.324 0.746 -0.006 0.008
TimeFocusing__MM01 -0.0046 0.003 -1.368 0.171 -0.011 0.002
==========================================================================================OLS + 통제변수 + 기업 고정효과 + 연도 고정효과
CAR[-1,+1] | N=9,410 | 통제변수·기업·연도 고정효과 포함
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m1_p1 R-squared: 0.128
Model: OLS Adj. R-squared: 0.036
Method: Least Squares F-statistic: 1.395
Date: Tue, 22 Sep 2026 Prob (F-statistic): 1.02e-12
Time: 04:54:54 Log-Likelihood: 17651.
No. Observations: 9410 AIC: -3.351e+04
Df Residuals: 8512 BIC: -2.709e+04
Df Model: 897
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept -0.0366 0.039 -0.928 0.354 -0.114 0.041
AI_Mention 0.0004 0.001 0.299 0.765 -0.002 0.003
Concreteness_All__MM01 -0.0119 0.013 -0.933 0.351 -0.037 0.013
TimeFocusing__MM01 -0.0023 0.009 -0.259 0.796 -0.020 0.015
FOG__RAW 0.0006 0.001 0.725 0.468 -0.001 0.002
FileSize__LN -0.0013 0.001 -0.887 0.375 -0.004 0.002
WordCount__LN -0.0004 0.003 -0.152 0.879 -0.006 0.005
PositiveTone__RAW -0.0062 0.006 -0.980 0.327 -0.019 0.006
NegativeTone__RAW 0.0002 0.002 0.083 0.934 -0.004 0.004
LEV__RAW -0.0055 0.005 -1.212 0.225 -0.014 0.003
SIZE__LN_ASSETS -0.0036 0.001 -2.645 0.008 -0.006 -0.001
CASH__RAW -0.0014 0.007 -0.198 0.843 -0.016 0.013
==========================================================================================
CAR[-2,+2] | N=9,405 | 통제변수·기업·연도 고정효과 포함
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m2_p2 R-squared: 0.127
Model: OLS Adj. R-squared: 0.035
Method: Least Squares F-statistic: 1.384
Date: Tue, 22 Sep 2026 Prob (F-statistic): 3.53e-12
Time: 04:54:59 Log-Likelihood: 15884.
No. Observations: 9405 AIC: -2.997e+04
Df Residuals: 8507 BIC: -2.355e+04
Df Model: 897
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept -0.0529 0.048 -1.113 0.266 -0.146 0.040
AI_Mention 0.0009 0.002 0.528 0.598 -0.002 0.004
Concreteness_All__MM01 -0.0097 0.015 -0.630 0.528 -0.040 0.021
TimeFocusing__MM01 -0.0156 0.011 -1.465 0.143 -0.036 0.005
FOG__RAW -0.0002 0.001 -0.144 0.885 -0.002 0.002
FileSize__LN -0.0027 0.002 -1.573 0.116 -0.006 0.001
WordCount__LN 0.0037 0.003 1.170 0.242 -0.003 0.010
PositiveTone__RAW -0.0108 0.008 -1.406 0.160 -0.026 0.004
NegativeTone__RAW 0.0020 0.003 0.786 0.432 -0.003 0.007
LEV__RAW -0.0080 0.005 -1.456 0.146 -0.019 0.003
SIZE__LN_ASSETS -0.0045 0.002 -2.807 0.005 -0.008 -0.001
CASH__RAW -0.0050 0.009 -0.572 0.567 -0.022 0.012
==========================================================================================
CAR[-3,+3] | N=9,402 | 통제변수·기업·연도 고정효과 포함
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m3_p3 R-squared: 0.134
Model: OLS Adj. R-squared: 0.042
Method: Least Squares F-statistic: 1.463
Date: Tue, 22 Sep 2026 Prob (F-statistic): 3.32e-16
Time: 04:55:06 Log-Likelihood: 14455.
No. Observations: 9402 AIC: -2.711e+04
Df Residuals: 8504 BIC: -2.070e+04
Df Model: 897
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept 0.0971 0.055 1.755 0.079 -0.011 0.206
AI_Mention 0.0011 0.002 0.574 0.566 -0.003 0.005
Concreteness_All__MM01 -0.0197 0.018 -1.099 0.272 -0.055 0.015
TimeFocusing__MM01 -0.0061 0.012 -0.497 0.619 -0.030 0.018
FOG__RAW -0.0009 0.001 -0.718 0.473 -0.003 0.002
FileSize__LN -0.0032 0.002 -1.593 0.111 -0.007 0.001
WordCount__LN 0.0006 0.004 0.159 0.874 -0.007 0.008
PositiveTone__RAW -0.0072 0.009 -0.804 0.422 -0.025 0.010
NegativeTone__RAW 0.0011 0.003 0.353 0.724 -0.005 0.007
LEV__RAW -0.0094 0.006 -1.468 0.142 -0.022 0.003
SIZE__LN_ASSETS -0.0049 0.002 -2.611 0.009 -0.009 -0.001
CASH__RAW 4.388e-05 0.010 0.004 0.997 -0.020 0.020
==========================================================================================
CAR[-1,+2] | N=9,405 | 통제변수·기업·연도 고정효과 포함
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m1_p2 R-squared: 0.131
Model: OLS Adj. R-squared: 0.039
Method: Least Squares F-statistic: 1.429
Date: Tue, 22 Sep 2026 Prob (F-statistic): 2.11e-14
Time: 04:55:11 Log-Likelihood: 16690.
No. Observations: 9405 AIC: -3.158e+04
Df Residuals: 8507 BIC: -2.516e+04
Df Model: 897
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept -0.0686 0.044 -1.571 0.116 -0.154 0.017
AI_Mention 0.0002 0.001 0.120 0.904 -0.003 0.003
Concreteness_All__MM01 0.0030 0.014 0.214 0.831 -0.025 0.031
TimeFocusing__MM01 -0.0153 0.010 -1.570 0.116 -0.034 0.004
FOG__RAW 0.0003 0.001 0.282 0.778 -0.002 0.002
FileSize__LN -0.0016 0.002 -1.043 0.297 -0.005 0.001
WordCount__LN 0.0034 0.003 1.179 0.238 -0.002 0.009
PositiveTone__RAW -0.0112 0.007 -1.594 0.111 -0.025 0.003
NegativeTone__RAW 0.0021 0.002 0.884 0.377 -0.003 0.007
LEV__RAW -0.0095 0.005 -1.897 0.058 -0.019 0.000
SIZE__LN_ASSETS -0.0042 0.001 -2.796 0.005 -0.007 -0.001
CASH__RAW -0.0071 0.008 -0.887 0.375 -0.023 0.009
==========================================================================================
CAR[-2,0] | N=9,410 | 통제변수·기업·연도 고정효과 포함
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m2_0 R-squared: 0.123
Model: OLS Adj. R-squared: 0.031
Method: Least Squares F-statistic: 1.336
Date: Tue, 22 Sep 2026 Prob (F-statistic): 5.94e-10
Time: 04:55:18 Log-Likelihood: 17893.
No. Observations: 9410 AIC: -3.399e+04
Df Residuals: 8512 BIC: -2.757e+04
Df Model: 897
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept 0.0327 0.038 0.851 0.395 -0.043 0.108
AI_Mention -0.0008 0.001 -0.603 0.547 -0.003 0.002
Concreteness_All__MM01 -0.0231 0.012 -1.849 0.065 -0.048 0.001
TimeFocusing__MM01 -7.563e-05 0.009 -0.009 0.993 -0.017 0.017
FOG__RAW -0.0003 0.001 -0.341 0.733 -0.002 0.001
FileSize__LN -0.0017 0.001 -1.219 0.223 -0.004 0.001
WordCount__LN -0.0005 0.003 -0.176 0.860 -0.005 0.005
PositiveTone__RAW -0.0029 0.006 -0.462 0.644 -0.015 0.009
NegativeTone__RAW -0.0004 0.002 -0.204 0.838 -0.005 0.004
LEV__RAW -0.0030 0.004 -0.667 0.505 -0.012 0.006
SIZE__LN_ASSETS -0.0036 0.001 -2.748 0.006 -0.006 -0.001
CASH__RAW 3.831e-05 0.007 0.005 0.996 -0.014 0.014
==========================================================================================
CAR[0,+2] | N=9,405 | 통제변수·기업·연도 고정효과 포함
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_0_p2 R-squared: 0.138
Model: OLS Adj. R-squared: 0.047
Method: Least Squares F-statistic: 1.519
Date: Tue, 22 Sep 2026 Prob (F-statistic): 2.28e-19
Time: 04:55:23 Log-Likelihood: 18528.
No. Observations: 9405 AIC: -3.526e+04
Df Residuals: 8507 BIC: -2.884e+04
Df Model: 897
Covariance Type: nonrobust
==============================================================================
==========================================================================================
coef std err t P>|t| [0.025 0.975]
------------------------------------------------------------------------------------------
Intercept -0.0932 0.036 -2.594 0.009 -0.164 -0.023
AI_Mention 0.0014 0.001 1.118 0.264 -0.001 0.004
Concreteness_All__MM01 0.0182 0.012 1.558 0.119 -0.005 0.041
TimeFocusing__MM01 -0.0170 0.008 -2.118 0.034 -0.033 -0.001
FOG__RAW 0.0002 0.001 0.309 0.757 -0.001 0.002
FileSize__LN -0.0010 0.001 -0.809 0.419 -0.004 0.001
WordCount__LN 0.0049 0.002 2.058 0.040 0.000 0.010
PositiveTone__RAW -0.0062 0.006 -1.079 0.281 -0.018 0.005
NegativeTone__RAW 0.0027 0.002 1.357 0.175 -0.001 0.007
LEV__RAW -0.0066 0.004 -1.596 0.110 -0.015 0.002
SIZE__LN_ASSETS -0.0028 0.001 -2.302 0.021 -0.005 -0.000
CASH__RAW -0.0091 0.007 -1.396 0.163 -0.022 0.004
==========================================================================================AI 언급 시작
CAR[-1,+1] | N=8,886 | 전체 10-K: 신규 AI 언급(0→1)
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m1_p1 R-squared: 0.138
Model: OLS Adj. R-squared: 0.043
Method: Least Squares F-statistic: 1.455
Date: Tue, 22 Sep 2026 Prob (F-statistic): 1.77e-15
Time: 05:43:02 Log-Likelihood: 16809.
No. Observations: 8886 AIC: -3.185e+04
Df Residuals: 8002 BIC: -2.558e+04
Df Model: 883
Covariance Type: nonrobust
==============================================================================
=========================================================================================
coef std err t P>|t| [0.025 0.975]
-----------------------------------------------------------------------------------------
Intercept 0.0448 0.032 1.379 0.168 -0.019 0.108
AI_Mention_Start_0to1 -0.0046 0.002 -2.494 0.013 -0.008 -0.001
AI_Mention_Stop_1to0 -0.0007 0.002 -0.293 0.769 -0.006 0.004
FOG__RAW 0.0007 0.001 0.796 0.426 -0.001 0.002
FileSize__LN -0.0009 0.001 -0.639 0.523 -0.004 0.002
WordCount__LN 0.0001 0.002 0.056 0.955 -0.005 0.005
PositiveTone__RAW -0.0068 0.006 -1.086 0.277 -0.019 0.005
NegativeTone__RAW 0.0005 0.002 0.255 0.799 -0.004 0.005
LEV__RAW -0.0051 0.005 -1.079 0.281 -0.014 0.004
SIZE__LN_ASSETS -0.0041 0.001 -2.884 0.004 -0.007 -0.001
CASH__RAW -0.0020 0.007 -0.264 0.792 -0.016 0.013
=========================================================================================
CAR[-2,+2] | N=8,881 | 전체 10-K: 신규 AI 언급(0→1)
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m2_p2 R-squared: 0.139
Model: OLS Adj. R-squared: 0.044
Method: Least Squares F-statistic: 1.468
Date: Tue, 22 Sep 2026 Prob (F-statistic): 3.67e-16
Time: 05:43:09 Log-Likelihood: 15156.
No. Observations: 8881 AIC: -2.854e+04
Df Residuals: 7997 BIC: -2.227e+04
Df Model: 883
Covariance Type: nonrobust
==============================================================================
=========================================================================================
coef std err t P>|t| [0.025 0.975]
-----------------------------------------------------------------------------------------
Intercept 0.0373 0.039 0.953 0.341 -0.039 0.114
AI_Mention_Start_0to1 -0.0049 0.002 -2.203 0.028 -0.009 -0.001
AI_Mention_Stop_1to0 -0.0020 0.003 -0.664 0.506 -0.008 0.004
FOG__RAW -7.228e-05 0.001 -0.068 0.946 -0.002 0.002
FileSize__LN -0.0020 0.002 -1.161 0.246 -0.005 0.001
WordCount__LN 0.0031 0.003 1.096 0.273 -0.002 0.009
PositiveTone__RAW -0.0087 0.007 -1.163 0.245 -0.023 0.006
NegativeTone__RAW 0.0022 0.003 0.875 0.382 -0.003 0.007
LEV__RAW -0.0087 0.006 -1.536 0.125 -0.020 0.002
SIZE__LN_ASSETS -0.0045 0.002 -2.656 0.008 -0.008 -0.001
CASH__RAW -0.0092 0.009 -1.030 0.303 -0.027 0.008
=========================================================================================
CAR[-3,+3] | N=8,878 | 전체 10-K: 신규 AI 언급(0→1)
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m3_p3 R-squared: 0.143
Model: OLS Adj. R-squared: 0.048
Method: Least Squares F-statistic: 1.507
Date: Tue, 22 Sep 2026 Prob (F-statistic): 2.56e-18
Time: 05:43:13 Log-Likelihood: 13772.
No. Observations: 8878 AIC: -2.578e+04
Df Residuals: 7994 BIC: -1.951e+04
Df Model: 883
Covariance Type: nonrobust
==============================================================================
=========================================================================================
coef std err t P>|t| [0.025 0.975]
-----------------------------------------------------------------------------------------
Intercept 0.0915 0.046 2.004 0.045 0.002 0.181
AI_Mention_Start_0to1 -0.0039 0.003 -1.518 0.129 -0.009 0.001
AI_Mention_Stop_1to0 -0.0020 0.003 -0.576 0.565 -0.009 0.005
FOG__RAW -0.0009 0.001 -0.684 0.494 -0.003 0.002
FileSize__LN -0.0024 0.002 -1.189 0.234 -0.006 0.002
WordCount__LN 0.0016 0.003 0.482 0.630 -0.005 0.008
PositiveTone__RAW -0.0059 0.009 -0.673 0.501 -0.023 0.011
NegativeTone__RAW 0.0020 0.003 0.655 0.512 -0.004 0.008
LEV__RAW -0.0096 0.007 -1.445 0.148 -0.023 0.003
SIZE__LN_ASSETS -0.0055 0.002 -2.773 0.006 -0.009 -0.002
CASH__RAW -0.0045 0.010 -0.431 0.666 -0.025 0.016
=========================================================================================
CAR[-1,+2] | N=8,881 | 전체 10-K: 신규 AI 언급(0→1)
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m1_p2 R-squared: 0.144
Model: OLS Adj. R-squared: 0.050
Method: Least Squares F-statistic: 1.529
Date: Tue, 22 Sep 2026 Prob (F-statistic): 1.42e-19
Time: 05:43:19 Log-Likelihood: 15888.
No. Observations: 8881 AIC: -3.001e+04
Df Residuals: 7997 BIC: -2.374e+04
Df Model: 883
Covariance Type: nonrobust
==============================================================================
=========================================================================================
coef std err t P>|t| [0.025 0.975]
-----------------------------------------------------------------------------------------
Intercept 0.0353 0.036 0.980 0.327 -0.035 0.106
AI_Mention_Start_0to1 -0.0045 0.002 -2.188 0.029 -0.008 -0.000
AI_Mention_Stop_1to0 -0.0006 0.003 -0.204 0.838 -0.006 0.005
FOG__RAW 0.0002 0.001 0.163 0.871 -0.002 0.002
FileSize__LN -0.0014 0.002 -0.899 0.368 -0.004 0.002
WordCount__LN 0.0027 0.003 1.010 0.312 -0.003 0.008
PositiveTone__RAW -0.0101 0.007 -1.460 0.144 -0.024 0.003
NegativeTone__RAW 0.0016 0.002 0.685 0.493 -0.003 0.006
LEV__RAW -0.0105 0.005 -2.001 0.045 -0.021 -0.000
SIZE__LN_ASSETS -0.0045 0.002 -2.840 0.005 -0.008 -0.001
CASH__RAW -0.0082 0.008 -1.006 0.314 -0.024 0.008
=========================================================================================
CAR[-2,0] | N=8,886 | 전체 10-K: 신규 AI 언급(0→1)
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_m2_0 R-squared: 0.132
Model: OLS Adj. R-squared: 0.036
Method: Least Squares F-statistic: 1.376
Date: Tue, 22 Sep 2026 Prob (F-statistic): 1.36e-11
Time: 05:43:24 Log-Likelihood: 17122.
No. Observations: 8886 AIC: -3.248e+04
Df Residuals: 8002 BIC: -2.621e+04
Df Model: 883
Covariance Type: nonrobust
==============================================================================
=========================================================================================
coef std err t P>|t| [0.025 0.975]
-----------------------------------------------------------------------------------------
Intercept 0.0466 0.031 1.487 0.137 -0.015 0.108
AI_Mention_Start_0to1 -0.0045 0.002 -2.558 0.011 -0.008 -0.001
AI_Mention_Stop_1to0 -0.0006 0.002 -0.257 0.797 -0.005 0.004
FOG__RAW -2.475e-05 0.001 -0.029 0.977 -0.002 0.002
FileSize__LN -0.0008 0.001 -0.619 0.536 -0.004 0.002
WordCount__LN -7.247e-05 0.002 -0.031 0.975 -0.005 0.004
PositiveTone__RAW -0.0017 0.006 -0.276 0.782 -0.013 0.010
NegativeTone__RAW 0.0002 0.002 0.075 0.941 -0.004 0.004
LEV__RAW -0.0026 0.005 -0.575 0.565 -0.012 0.006
SIZE__LN_ASSETS -0.0038 0.001 -2.764 0.006 -0.006 -0.001
CASH__RAW -0.0034 0.007 -0.482 0.630 -0.017 0.011
=========================================================================================
CAR[0,+2] | N=8,881 | 전체 10-K: 신규 AI 언급(0→1)
OLS Regression Results
==============================================================================
Dep. Variable: V8_CAR_MM_0_p2 R-squared: 0.157
Model: OLS Adj. R-squared: 0.064
Method: Least Squares F-statistic: 1.691
Date: Tue, 22 Sep 2026 Prob (F-statistic): 1.10e-29
Time: 05:43:28 Log-Likelihood: 17658.
No. Observations: 8881 AIC: -3.355e+04
Df Residuals: 7997 BIC: -2.728e+04
Df Model: 883
Covariance Type: nonrobust
==============================================================================
=========================================================================================
coef std err t P>|t| [0.025 0.975]
-----------------------------------------------------------------------------------------
Intercept 0.0061 0.029 0.207 0.836 -0.052 0.064
AI_Mention_Start_0to1 -0.0027 0.002 -1.635 0.102 -0.006 0.001
AI_Mention_Stop_1to0 -0.0015 0.002 -0.659 0.510 -0.006 0.003
FOG__RAW 2.912e-05 0.001 0.036 0.971 -0.002 0.002
FileSize__LN -0.0013 0.001 -0.989 0.323 -0.004 0.001
WordCount__LN 0.0037 0.002 1.704 0.088 -0.001 0.008
PositiveTone__RAW -0.0055 0.006 -0.966 0.334 -0.017 0.006
NegativeTone__RAW 0.0017 0.002 0.895 0.371 -0.002 0.006
LEV__RAW -0.0069 0.004 -1.618 0.106 -0.015 0.001
SIZE__LN_ASSETS -0.0029 0.001 -2.219 0.026 -0.005 -0.000
CASH__RAW -0.0070 0.007 -1.044 0.296 -0.020 0.006
=========================================================================================