Creating Effective Revenue Forecast Models for CCAR
Creating Effective Revenue Forecast Models for CCAR Michael Woods, Principal Data Scientist
1
Proprietary
To see this content presented, view the webinar at: https://youtu.be/8pnpdrSis7w
2
Proprietary
Complexity is the Challenge
# of possible insights in the Dataset is exponential in size
c
Dataset 3
Proprietary
# of possible insights in the Dataset next year after it grows another 40%
Inadequate Response to Complexity Quants
Conventional Analytical Approaches
Overfitting y
x
4
Proprietary
Clean Slate Requirements
Speed
5
Proprietary
Accuracy
Defensibility
A Man-Machine Workflow
Data
Algorithms + Compute
Variable Identification Machines
6
Proprietary
Group of Variables
Business Input
Variable Selection
Statistical Tests
Models
Model Identification
Business Validation
Model Selection
Machine Intelligence Scalable Compute
Machine Learning, Geometric + Statistical Algorithms
+
7
Proprietary
+
Topological Data Analysis
Transforming Data to Insight
Fed Macro Variables ~300
8
Proprietary
Transforms ~900
Lagged ~2700
Distillable Model
Correlate Variables with Business Performance
Correlation with Revenue High
9
Proprietary
Low
EurozoneUnemployment 20-Bond GO Index_Lag_1
UKGB10Yr
Corr AbsCorr 0.78 0.78 0.71 0.71 0.67 0.67 -0.66 0.66 0.62 0.62 0.51 0.51
B
A
Group B
Eurozone Interest Rates Eurozonei1MLibor UKi1MLibor Eurozonei5YrSwap
10
Proprietary
Corr AbsCorr 0.85 0.85 0.76 0.76 0.67 0.67 -0.34 0.34
Group C
D Group D
Funding Stress TED LIBOR less OIS LIBOR_Less_FED Eurozonei1MLibor_delta
Macro Growth EurozoneGDP_Index UKGDP_Index USUnemployment UKHousing_Lag_2
Corr AbsCorr 0.66 0.66 0.59 0.59 -0.57 0.57 0.47 0.47
C
E
Group E
Group A
Identify Statistically Significant Variables
Emerging Risk ASIAHangseng MexicoFxrate EMEAMXEMEA_Lag_2
UKFxrate
Credit Spreads USCDS_US10YR_BBB_pctChange USCDXnonIG_pctChange USCDS_US10YR_A_pctChange USCDS_US25YR_BBB_pctChange USCDXIG_pctChange
Corr AbsCorr -0.67 0.67 -0.51 0.51 0.49 0.49 -0.48 0.48 Corr AbsCorr 0.70 0.70 0.70 0.70 0.68 0.68 0.65 0.65 0.58 0.58
EurozoneUnemployment 20-Bond GO Index_Lag_1
UKGB10Yr
Corr AbsCorr 0.78 0.78 0.71 0.71 0.67 0.67 -0.66 0.66 0.62 0.62 0.51 0.51
B
Group B
Eurozone Interest Rates Eurozonei1MLibor UKi1MLibor Eurozonei5YrSwap
Macro Growth EurozoneGDP_Index UKGDP_Index USUnemployment UKHousing_Lag_2
Corr AbsCorr 0.66 0.66 0.59 0.59 -0.57 0.57 0.47 0.47
A
E
11
Funding Stress TED LIBOR less OIS LIBOR_Less_FED Eurozonei1MLibor_delta
Proprietary
Corr AbsCorr 0.85 0.85 0.76 0.76 0.67 0.67 -0.34 0.34
D Group D
Group E
Group A
Identify Statistically Significant Variables
Credit Spreads USCDS_US10YR_BBB_pctChange USCDXnonIG_pctChange USCDS_US10YR_A_pctChange USCDS_US25YR_BBB_pctChange USCDXIG_pctChange
Corr AbsCorr 0.70 0.70 0.70 0.70 0.68 0.68 0.65 0.65 0.58 0.58
Generate Permutation Set of Models
12
Proprietary
Select for Empirical Validity
13
Proprietary
Identify the Champion Model
14
Proprietary
A Man-Machine Workflow
Data
Algorithms + Compute
Variable Identification Machines
15
Proprietary
Group of Variables
Business Input
Variable Selection
Statistical Tests
Models
Model Identification
Business Validation
Model Selection
4400 Bohannon Drive, Suite 200 Menlo Park, CA 94025 ayasdi.com | @ayasdi
1
Proprietary
To see this content presented, view the webinar at: https://youtu.be/8pnpdrSis7w
2
Proprietary
Complexity is the Challenge
# of possible insights in the Dataset is exponential in size
c
Dataset 3
Proprietary
# of possible insights in the Dataset next year after it grows another 40%
Inadequate Response to Complexity Quants
Conventional Analytical Approaches
Overfitting y
x
4
Proprietary
Clean Slate Requirements
Speed
5
Proprietary
Accuracy
Defensibility
A Man-Machine Workflow
Data
Algorithms + Compute
Variable Identification Machines
6
Proprietary
Group of Variables
Business Input
Variable Selection
Statistical Tests
Models
Model Identification
Business Validation
Model Selection
Machine Intelligence Scalable Compute
Machine Learning, Geometric + Statistical Algorithms
+
7
Proprietary
+
Topological Data Analysis
Transforming Data to Insight
Fed Macro Variables ~300
8
Proprietary
Transforms ~900
Lagged ~2700
Distillable Model
Correlate Variables with Business Performance
Correlation with Revenue High
9
Proprietary
Low
EurozoneUnemployment 20-Bond GO Index_Lag_1
UKGB10Yr
Corr AbsCorr 0.78 0.78 0.71 0.71 0.67 0.67 -0.66 0.66 0.62 0.62 0.51 0.51
B
A
Group B
Eurozone Interest Rates Eurozonei1MLibor UKi1MLibor Eurozonei5YrSwap
10
Proprietary
Corr AbsCorr 0.85 0.85 0.76 0.76 0.67 0.67 -0.34 0.34
Group C
D Group D
Funding Stress TED LIBOR less OIS LIBOR_Less_FED Eurozonei1MLibor_delta
Macro Growth EurozoneGDP_Index UKGDP_Index USUnemployment UKHousing_Lag_2
Corr AbsCorr 0.66 0.66 0.59 0.59 -0.57 0.57 0.47 0.47
C
E
Group E
Group A
Identify Statistically Significant Variables
Emerging Risk ASIAHangseng MexicoFxrate EMEAMXEMEA_Lag_2
UKFxrate
Credit Spreads USCDS_US10YR_BBB_pctChange USCDXnonIG_pctChange USCDS_US10YR_A_pctChange USCDS_US25YR_BBB_pctChange USCDXIG_pctChange
Corr AbsCorr -0.67 0.67 -0.51 0.51 0.49 0.49 -0.48 0.48 Corr AbsCorr 0.70 0.70 0.70 0.70 0.68 0.68 0.65 0.65 0.58 0.58
EurozoneUnemployment 20-Bond GO Index_Lag_1
UKGB10Yr
Corr AbsCorr 0.78 0.78 0.71 0.71 0.67 0.67 -0.66 0.66 0.62 0.62 0.51 0.51
B
Group B
Eurozone Interest Rates Eurozonei1MLibor UKi1MLibor Eurozonei5YrSwap
Macro Growth EurozoneGDP_Index UKGDP_Index USUnemployment UKHousing_Lag_2
Corr AbsCorr 0.66 0.66 0.59 0.59 -0.57 0.57 0.47 0.47
A
E
11
Funding Stress TED LIBOR less OIS LIBOR_Less_FED Eurozonei1MLibor_delta
Proprietary
Corr AbsCorr 0.85 0.85 0.76 0.76 0.67 0.67 -0.34 0.34
D Group D
Group E
Group A
Identify Statistically Significant Variables
Credit Spreads USCDS_US10YR_BBB_pctChange USCDXnonIG_pctChange USCDS_US10YR_A_pctChange USCDS_US25YR_BBB_pctChange USCDXIG_pctChange
Corr AbsCorr 0.70 0.70 0.70 0.70 0.68 0.68 0.65 0.65 0.58 0.58
Generate Permutation Set of Models
12
Proprietary
Select for Empirical Validity
13
Proprietary
Identify the Champion Model
14
Proprietary
A Man-Machine Workflow
Data
Algorithms + Compute
Variable Identification Machines
15
Proprietary
Group of Variables
Business Input
Variable Selection
Statistical Tests
Models
Model Identification
Business Validation
Model Selection
4400 Bohannon Drive, Suite 200 Menlo Park, CA 94025 ayasdi.com | @ayasdi
