MPdist is a recently introduced distance measure which considers two time series to be similar if they share many similar subsequences, regardless of the order of matching subsequences. It was demonstrated in that MPdist is robust to spikes, warping, linear trends, dropouts, wandering baseline and missing values, issues that are common outside of benchmark datasets.
mpdist( ref_data, query_data, window_size, type = c("simple", "vector"), thr = 0.05 )
an int. Size of the sliding window.
the type of result. (Default is
threshold for MPdist. (Default is
Returns the distance of two time series or a vector containing the distance between all sliding windows.
MPdist returns the distance of two time series or a vector containing the distance
between all sliding windows. If argument
type is set to
vector, the vector is returned.
Gharghabi S, Imani S, Bagnall A, Darvishzadeh A, Keogh E. Matrix Profile XII: MPdist: A Novel Time Series Distance Measure to Allow Data Mining in More Challenging Scenarios. In: 2018 IEEE International Conference on Data Mining (ICDM). 2018.
ref_data <- mp_toy_data$data[, 1] qe_data <- mp_toy_data$data[, 2] qd_data <- mp_toy_data$data[150:200, 1] w <- mp_toy_data$sub_len # distance between data of same size deq <- mpdist(ref_data, qe_data, w) # distance between data of different sizes ddiff <- mpdist(ref_data, qd_data, w) #> Error: Time series is too short relative to desired window size. # distance vector between data of different sizes ddvect <- mpdist(ref_data, qd_data, w, type = "vector") #> Error in convert_data(data): Argument `data` cannot be `matrix` and must have a single dimension.Argument `data` cannot be `array` and must have a single dimension.