OwlMix

OwlMin Exploratory Data Analysis Report

Report generated on 2026-05-23 22:34:14

📊 ACF & PACF Analysis

The ACF and PACF tables present the numerical values of autocorrelation and partial autocorrelation coefficients for each variable across specified lags. These tabulated values enable precise identification of statistically significant lags and the strength of relationships over time.

		Lag
Columns	Type	0	1	2	3	4	5	6	7	8	9	10
radio_spend	ACF	1.000	0.040	-0.040	0.020	-0.020	0.040	-0.100	0.010	-0.010	-0.040	-0.020
radio_spend	PACF	1.000	0.040	-0.040	0.020	-0.030	0.050	-0.110	0.030	-0.020	-0.030	-0.030


digital_imp	ACF	1.000	-0.050	0.030	-0.020	-0.040	-0.040	-0.050	-0.050	0.050	0.060	0.000
digital_imp	PACF	1.000	-0.050	0.030	-0.020	-0.040	-0.040	-0.050	-0.060	0.040	0.060	-0.000


sales	ACF	1.000	-0.100	-0.110	0.090	-0.060	0.000	0.070	-0.040	0.060	-0.030	-0.030
sales	PACF	1.000	-0.100	-0.120	0.060	-0.060	0.010	0.060	-0.010	0.070	-0.040	-0.010

📈 Variance Inflation Factor (VIF) Analysis

The VIF table presents the variance inflation factors for each variable, helping to identify multicollinearity in your dataset.

Feature Variable	VIF Value	Chart
tv_spend	12.830
inflation	11.280
digital_spend	10.990
radio_grp	10.900
radio_imp	10.740
digital_imp	9.900
tv_grp	9.590
radio_spend	9.190

🔗 Correlation Analysis

The correlation analysis section provides insights into the relationships between variables in your dataset. It includes both the correlation matrix and the lagged correlation matrix, along with visualizations to help you understand these relationships better.

Correlation Matrix

	tv_spend	digital_spend	radio_spend	tv_grp	radio_grp	digital_imp	radio_imp	inflation	sales
tv_spend	1.00	-0.03	0.06	-0.01	-0.04	-0.08	-0.03	0.08	0.56
digital_spend	-0.03	1.00	0.04	0.07	0.04	-0.06	-0.03	-0.01	0.63
radio_spend	0.06	0.04	1.00	-0.02	-0.04	-0.04	0.02	0.08	0.15
tv_grp	-0.01	0.07	-0.02	1.00	0.02	-0.10	0.09	-0.07	0.09
radio_grp	-0.04	0.04	-0.04	0.02	1.00	0.12	0.04	0.07	0.09
digital_imp	-0.08	-0.06	-0.04	-0.10	0.12	1.00	0.04	0.05	0.01
radio_imp	-0.03	-0.03	0.02	0.09	0.04	0.04	1.00	0.03	-0.14
inflation	0.08	-0.01	0.08	-0.07	0.07	0.05	0.03	1.00	0.05
sales	0.56	0.63	0.15	0.09	0.09	0.01	-0.14	0.05	1.00

Lagged Correlation Matrix

Column	Lag 0	Lag 1	Lag 2	Lag 3	Lag 4	Lag 5	Lag 6	Lag 7	Lag 8
tv_spend	1.00	-0.12	-0.06	-0.01	-0.12	0.04	0.06	-0.04	0.02
digital_spend	1.00	-0.11	-0.01	0.10	-0.01	0.03	0.03	0.03	-0.03
radio_spend	1.00	0.04	-0.04	0.02	-0.02	0.04	-0.10	0.01	-0.01
tv_grp	1.00	0.03	0.06	0.05	0.04	-0.08	0.13	0.03	0.03
radio_grp	1.00	-0.01	0.10	0.02	0.02	-0.08	-0.08	-0.03	-0.05
digital_imp	1.00	-0.05	0.03	-0.02	-0.04	-0.04	-0.05	-0.05	0.05
radio_imp	1.00	-0.01	0.00	-0.03	0.00	0.06	-0.03	-0.05	-0.01
inflation	1.00	-0.04	0.01	-0.02	0.02	0.04	-0.05	-0.04	0.06
sales	1.00	-0.10	-0.11	0.09	-0.07	0.00	0.07	-0.04	0.06

Visualizations Heatmap for Correlation Analysis

🔗 Causality Analysis

The causality analysis section summarizes the results of causality tests between variables and the target. It includes key statistics such as best lag, p-values, scores, and whether a variable is considered causal.

Error Threshold: 15.0 %

Variable	Best Lag	P-Value	Min P-Value	Score	MAPE Score	Lags Tested	Causal	Coeff. Sign
tv_spend	1	0.274	0.274	78.45	12.832	5	No	Negative
digital_spend	1	0.436	0.436	69.18	11.680	5	No	Negative
radio_spend	2	0.538	0.538	61.55	15.371	5	No	Negative
tv_grp	5	0.471	0.471	65.47	15.672	5	No	Negative
radio_grp	1	0.010	0.010	93.17	15.552	5	No	Negative
digital_imp	5	0.233	0.233	79.77	15.667	5	No	Negative
radio_imp	1	0.004	0.004	93.58	15.495	5	No	Negative
inflation	2	0.481	0.481	64.87	15.627	5	No	Negative

📦 Box Plot Analysis

The box plot table presents the summary statistics (min, Q1, median, Q3, max, outliers) for each variable. This helps visualize the distribution and identify potential outliers.

Column	Min	Q1	Median	Q3	Max	Outlier Count	Outliers
tv_spend	104	241	303	348	483	23	143.37, 131.83, 440.65, 157.47, 482.9, 142.7, 454.15, 443.17, 144.06, 467.67, 161.75, 129.24, 159.3, 442.13, 120.83, 427.93, 432.69, 457.98, 134.15, 149.07, 104.01, 121.95, 156.4
digital_spend	51	134	167	197	283	33	280.77, 72.94, 257.29, 59.5, 268.17, 279.25, 77.73, 64.7, 80.61, 78.34, 51.22, 282.79, 262.24, 254.52, 61.69, 80.53, 64.71, 259.04, 251.03, 75.51, 81.55, 257.03, 251.7, 61.93, 269.08, 74.57, 262.61, 72.59, 81.51, 251.47, 81.74, 277.84, 72.56
radio_spend	20	61	80	100	146	26	141.22, 31.63, 28.62, 129.1, 129.45, 20.23, 28.57, 135.17, 135.59, 142.59, 33.67, 141.8, 34.47, 28.08, 140.89, 132.52, 25.78, 146.09, 131.0, 28.7, 30.13, 135.35, 29.51, 30.66, 21.88, 140.04
tv_grp	15	41	55	67	98	32	20.41, 87.9, 21.95, 22.51, 92.71, 14.59, 17.33, 86.73, 90.93, 20.3, 88.21, 84.38, 21.61, 23.02, 98.04, 85.8, 21.92, 90.68, 20.81, 91.52, 23.44, 22.04, 19.9, 22.96, 23.2, 23.88, 22.32, 85.09, 17.79, 90.53, 97.07, 21.74
digital_imp	11	45	54	67	100	28	10.57, 84.8, 93.77, 15.15, 88.6, 90.09, 96.14, 21.21, 94.75, 94.07, 88.17, 25.26, 86.05, 23.02, 17.25, 97.18, 24.09, 24.01, 89.05, 99.95, 23.04, 23.51, 86.32, 18.9, 18.65, 21.32, 12.27, 91.74

🔗 Cross-Correlation Function (CCF) Summary Table

The CCF summary table presents cross-correlation statistics between variable pairs, helping to identify lagged relationships.

Target	Feature	Version	Max-Corr	Max-Corr Lag	Corr@Lag0
sales	tv_spend	original	0.563	0	0.563
sales	tv_spend	adstock	0.479	0	0.479
sales	tv_spend	difference	0.396	0	0.396
sales	tv_spend	lag	-0.104	1	-0.031
sales	digital_spend	original	0.627	0	0.627
sales	digital_spend	adstock	0.521	0	0.521
sales	digital_spend	difference	0.48	0	0.480
sales	digital_spend	lag	-0.088	0	-0.088

📈 〜 Response Curve and Metrics

The response curve and metrics section provides a comprehensive overview of how each feature in your dataset responds to changes in spend. For each feature, you'll find key metrics such as current spend, average spend, RoI, saturation point, efficiency ratio, and status. Additionally, visualizations of the response curve and marginal RoI are included to help you understand the effectiveness of your investments and identify opportunities for optimization.

Note: This is not a business recommendation but rather an analytical tool to guide your decision-making process.

Feature	Metrics	Response Curve	Marginal RoI
digital_imp	Current Spend: 100 Average Spend: 55 RoI: 0.65 Saturation Point: 101.35 Efficiency Ratio: 0.20 Status: saturated
digital_spend	Current Spend: 283 Average Spend: 168 RoI: 1.36 Saturation Point: 124.06 Efficiency Ratio: 0.02 Status: saturated
radio_spend	Current Spend: 146 Average Spend: 83 RoI: 0.35 Saturation Point: 100.91 Efficiency Ratio: 0.08 Status: saturated
tv_grp	Current Spend: 98 Average Spend: 56 RoI: 0.37 Saturation Point: 101.18 Efficiency Ratio: 0.21 Status: saturated
tv_spend	Current Spend: 483 Average Spend: 295 RoI: 0.83 Saturation Point: 202.80 Efficiency Ratio: 0.02 Status: saturated

📊 〜 Time Series Decomposition

This section shows decomposition of the target series into observed, trend, seasonal, and residual components.

Description	Chart
Actual (Observed) KPI values over time before decomposition.	$Observed series chart$
Trend component showing the underlying trend in the data over time.
Seasonal component showing repeating patterns or cycles in the data over time.
Unexplained residuals after removing trend and seasonal components, indicating noise or irregular patterns in the data.