la ciencia de datos, como disciplina, consta de dos partes:
## Posterior inference {CausalImpact}
##
## Average Cumulative
## Actual 117 3511
## Prediction (s.d.) 107 (0.35) 3196 (10.38)
## 95% CI [106, 107] [3176, 3217]
##
## Absolute effect (s.d.) 11 (0.35) 316 (10.38)
## 95% CI [9.8, 11] [294.4, 336]
##
## Relative effect (s.d.) 9.9% (0.32%) 9.9% (0.32%)
## 95% CI [9.2%, 11%] [9.2%, 11%]
##
## Posterior tail-area probability p: 0.00101
## Posterior prob. of a causal effect: 99.8994%
##
## For more details, type: summary(impact, "report")
During the post-intervention period, the response variable had an average value of approx. 117.05. By contrast, in the absence of an intervention, we would have expected an average response of 106.53. The 95% interval of this counterfactual prediction is [105.85, 107.23]. Subtracting this prediction from the observed response yields an estimate of the causal effect the intervention had on the response variable. This effect is 10.52 with a 95% interval of [9.81, 11.20]. For a discussion of the significance of this effect, see below. […]
se quiere medir el impacto causal del impuesto sobre varios indicadores:
## Posterior inference {CausalImpact}
##
## Average Cumulative
## Actual 0.88 26.36
## Prediction (s.d.) 1 (0.044) 30 (1.308)
## 95% CI [0.93, 1.1] [27.92, 32.7]
##
## Absolute effect (s.d.) -0.14 (0.044) -4.07 (1.308)
## 95% CI [-0.21, -0.052] [-6.36, -1.560]
##
## Relative effect (s.d.) -13% (4.3%) -13% (4.3%)
## 95% CI [-21%, -5.1%] [-21%, -5.1%]
##
## Posterior tail-area probability p: 0.00201
## Posterior prob. of a causal effect: 99.79879%
##
## For more details, type: summary(impact, "report")