0
$\begingroup$

Say I have a CausalImpact Summary like this:

> summary(impact)
Posterior inference {CausalImpact}

                         Average        Cumulative    
Actual                   3              65            
Prediction (s.d.)        2.7 (0.46)     60.4 (10.21)  
95% CI                   [1.8, 3.6]     [40.4, 79.6]  

Absolute effect (s.d.)   0.21 (0.46)    4.56 (10.21)  
95% CI                   [-0.66, 1.1]   [-14.63, 24.6]

Relative effect (s.d.)   7.5% (17%)     7.5% (17%)    
95% CI                   [-24%, 41%]    [-24%, 41%]   

Posterior tail-area probability p:   0.32852
Posterior prob. of a causal effect:  67%

For more details, type: summary(impact, "report")

How would I extract the Actual Average of 3? How would I extract the Posterior tail-area probability p?

I've tried summarizing numbers from the impact$series dataframe, but the average of the actual values after the date of intervention doesn't equal 3, so I'm stumped.

Thanks!

$\endgroup$
  • $\begingroup$ I guess you can find all you need in impact$summary. $\endgroup$ – user81847 Jan 19 '16 at 4:57
2
$\begingroup$

Using the first example of ?CausalImpact

set.seed(1)
x1 <- 100 + arima.sim(model = list(ar = 0.999), n = 100)
y <- 1.2 * x1 + rnorm(100)
y[71:100] <- y[71:100] + 10
data <- cbind(y, x1)
pre.period <- c(1, 70)
post.period <- c(71, 100)
impact <- CausalImpact(data, pre.period, post.period)

summary(impact)
# Posterior inference {CausalImpact}

#                          Average        Cumulative  
# Actual                   117            3511        
# Prediction (s.d.)        107 (0.42)     3195 (12.51)
# 95% CI                   [106, 107]     [3169, 3218]

# Absolute effect (s.d.)   11 (0.42)      316 (12.51) 
# 95% CI                   [9.8, 11]      [293.0, 342]

# Relative effect (s.d.)   9.9% (0.39%)   9.9% (0.39%)
# 95% CI                   [9.2%, 11%]    [9.2%, 11%] 

# Posterior tail-area probability p:   0.00111
# Posterior prob. of a causal effect:  99.88901%

# For more details, type: summary(impact, "report")

In this example Average for Actual is 117 and Posterior tail-area probability p is 0.00111.

impact object is a list of 4 objects (str(impact)). We are interested in the object impact$summary:

impact$summary["Average","Actual"]
# [1] 117.0485

impact$summary["Average","p"]
# [1] 0.001109878
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.