Short time serie error

[ddjustina]

Hello there,

I'm trying to apply the anomaly detection function to my time serie. Which is composed by only 60 observations, and it has 2 periods.

This is how I set up in the first place:

res <- AnomalyDetectionVec(data$value, max_anoms=0.4, period=29, 
                                           direction='neg', only_last=FALSE, 
                                           plot=TRUE)

# This is the output
$anoms
data frame with 0 columns and 0 rows
$plot
NULL

I don't know how, but I was able to run it in the first time. But now, I can't even get the graph, all the output variables are NULL. I already checked if the object's class was compatible, and it matches the same the class as the dataset used in the example ("raw_data").

This is the data: vector_data.txt

Diego Della Justina, PhD

[marchezinixd]

Did you solved this? I'm with the same problem here.

[ddjustina]

Unfortunately I did not. Actually I applied a filter to my time series instead

---

From: Guilherme Fernandes Marchezini notifications@github.com Sent: Wednesday, April 26, 2017 5:07:36 PM To: twitter/AnomalyDetection Cc: Diego Della Justina; Author Subject: Re: [twitter/AnomalyDetection] Short time serie error (#72)

Did you solved this? I'm with the same problem here.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/twitter/AnomalyDetection/issues/72#issuecomment-297526099, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AHnyCEFcn1nQId8uzDV-gAATMDYWVQEhks5rz6QIgaJpZM4IrLBQ.

[marchezinixd]

Thanks. I'm trying to use it with 254 samples, but i'm pretty sure it does not have seasonality.

[ddjustina]

I'm pretty sure that lack of seasonality is the cause of this problem. In my case I had seasonality, but my time series was very short, so I only had two sine waves at each sample, certainly not enough for the algorighm to learn. I only insisted, because it worked perfectly once, but when I runned my script again, It was no longer detecting the anomaly. I should have set the seed, that would probably return the initial result.

---

From: Guilherme Fernandes Marchezini notifications@github.com Sent: Thursday, April 27, 2017 1:36:20 PM To: twitter/AnomalyDetection Cc: Diego Della Justina; Author Subject: Re: [twitter/AnomalyDetection] Short time serie error (#72)

Thanks. I'm trying to use it with 254 samples, but i'm pretty sure it does not have seasonality.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/twitter/AnomalyDetection/issues/72#issuecomment-297770057, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AHnyCNoSEJDkmxIZ1stLFknREZldrID-ks5r0MQEgaJpZM4IrLBQ.

[marchezinixd]

Intresting, yes, the seed probably affected the result . Yes, problably the lack of seasonality is the main cause in my case. Thanks for the help.

[ddjustina]

you are welcome.

---

From: Guilherme Fernandes Marchezini notifications@github.com Sent: Thursday, April 27, 2017 3:07:37 PM To: twitter/AnomalyDetection Cc: Diego Della Justina; Author Subject: Re: [twitter/AnomalyDetection] Short time serie error (#72)

Intresting, yes, the seed probably affected the result . Yes, problably the lack of seasonality is the main cause in my case. Thanks for the help.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/twitter/AnomalyDetection/issues/72#issuecomment-297793960, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AHnyCP1uwAQk-KRdpELsB45YHl2Ea_dqks5r0NlpgaJpZM4IrLBQ.