Abstract
Predicting financial risk is a long-lasting challenge in consumer finance industry. Existing predictive models mainly rely on structured credit data but fail to work on unstructured non-financial data. A few studies that work on non-financial data usually focus on merely one data source. We follow a design science approach and propose a framework to predict financial risk within multiple non-financial data sources (i.e. within-app browsing behavior, short message, and customer social network). Based on the kernel theory of Predictive Analytics, we detail a design framework which first develops individual predictive models within each data domain and then ensembles them together for a multifaceted risk profiling. We conduct multiple experiments to evaluate the performance of this framework and find empirical support. This paper contributes to financial risk literature by proposing potential causal connections and to design science literature by demonstrating how to gain predictive power from various non-financial data sources.
| Original language | English (US) |
|---|---|
| State | Published - Jan 1 2019 |
| Event | 25th Americas Conference on Information Systems, AMCIS 2019 - Cancun, Mexico Duration: Aug 15 2019 → Aug 17 2019 |
Conference
| Conference | 25th Americas Conference on Information Systems, AMCIS 2019 |
|---|---|
| Country | Mexico |
| City | Cancun |
| Period | 8/15/19 → 8/17/19 |
Fingerprint
Keywords
- Behavioral data
- Design science
- Financial risk prediction
- Machine learning
- Non-financial data
- Predictive analytics
ASJC Scopus subject areas
- Information Systems
Cite this
Predicting financial risk using non-financial data : Design and evaluation of a predictive analytics framework. / Li, Chunxiao; Wang, Hongchang; Min, Wei; Tang, Zhengyang; Gu, Bin.
2019. Paper presented at 25th Americas Conference on Information Systems, AMCIS 2019, Cancun, Mexico.Research output: Contribution to conference › Paper
}
TY - CONF
T1 - Predicting financial risk using non-financial data
T2 - Design and evaluation of a predictive analytics framework
AU - Li, Chunxiao
AU - Wang, Hongchang
AU - Min, Wei
AU - Tang, Zhengyang
AU - Gu, Bin
PY - 2019/1/1
Y1 - 2019/1/1
N2 - Predicting financial risk is a long-lasting challenge in consumer finance industry. Existing predictive models mainly rely on structured credit data but fail to work on unstructured non-financial data. A few studies that work on non-financial data usually focus on merely one data source. We follow a design science approach and propose a framework to predict financial risk within multiple non-financial data sources (i.e. within-app browsing behavior, short message, and customer social network). Based on the kernel theory of Predictive Analytics, we detail a design framework which first develops individual predictive models within each data domain and then ensembles them together for a multifaceted risk profiling. We conduct multiple experiments to evaluate the performance of this framework and find empirical support. This paper contributes to financial risk literature by proposing potential causal connections and to design science literature by demonstrating how to gain predictive power from various non-financial data sources.
AB - Predicting financial risk is a long-lasting challenge in consumer finance industry. Existing predictive models mainly rely on structured credit data but fail to work on unstructured non-financial data. A few studies that work on non-financial data usually focus on merely one data source. We follow a design science approach and propose a framework to predict financial risk within multiple non-financial data sources (i.e. within-app browsing behavior, short message, and customer social network). Based on the kernel theory of Predictive Analytics, we detail a design framework which first develops individual predictive models within each data domain and then ensembles them together for a multifaceted risk profiling. We conduct multiple experiments to evaluate the performance of this framework and find empirical support. This paper contributes to financial risk literature by proposing potential causal connections and to design science literature by demonstrating how to gain predictive power from various non-financial data sources.
KW - Behavioral data
KW - Design science
KW - Financial risk prediction
KW - Machine learning
KW - Non-financial data
KW - Predictive analytics
UR - http://www.scopus.com/inward/record.url?scp=85073507147&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85073507147&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:85073507147
ER -