Table 2 Events characteristics depending on the caused stock price change and model performance metrics.

From: New drugs and stock market: a machine learning framework for predicting pharma market reaction to clinical trial announcements

Class name*

Extremely

Moderately

  

Moderately

Extremely

Negative

Negative

Negative

Positive

Positive

Positive

Stock price change range

(\(- \,\infty\), − 0.28]

(− 0.28, − 0.14]

(− 0.14, 0]

(0, 0.14]

(0.14, 0.28]

(0.28, + \(\infty\))

Number of events

211

189

599

478

110

67

Positive events**

72

106

421

366

83

57

Negative events**

139

83

178

112

27

10

OvR ROC AUC for GCN+GB

\(0.87 \pm 0.02\)

\(0.77 \pm 0.03\)

\(0.63 \pm 0.02\)

\(0.71 \pm 0.01\)

\(0.70 \pm 0.02\)

\(0.75 \pm 0.04\)

OvR ROC AUC for GB

\(0.85 \pm 0.02\)

\(0.72 \pm 0.02\)

\(0.60 \pm 0.02\)

\(0.67 \pm 0.02\)

\(0.66 \pm 0.04\)

\(0.74 \pm 0.05\)

Welch’s t-test p-value***

0.09

0.05

0.002

\(5.4 \times 10^{-5}\)

0.02

0.65

  1. *According to the value of price change.
  2. **According to the sentiment polarity of the announcement.
  3. ***Welch’s t-test p-value for equality of GB and GCN+GB metrics distributions.