Overview

numspellR is a tool to detect numeric persistence (“spells”) and rigidity patterns in numeric data. It works with numeric vectors and data frames.

Installation

# Install from your local package
# install.packages("numspellR")
library(numspellR)

# example
df <- data.frame(
  time = 1:8,
  value = c(5, 5, 5, 5, 6, 6, 6, 7)
)

numspellr(df, lang = "indonesia")

##      structure avg_spell median_spell max_spell stability_ratio
## 1 seri_numerik       3.5          3.5         4           0.875
##   elasticity_index adjustment_frequency spell_concentration persistence_status
## 1       0.05238095                 0.25           0.5102041        Sangat Kaku
##                                                                                                                                                                                                                                                                                     interpretation
## 1 Rata-rata nilai bertahan sekitar 3.5 periode sebelum mengalami perubahan.   Perubahan cukup sering terjadi.  Perubahan terjadi secara bertahap.  Periode stagnan terkonsentrasi pada beberapa fase panjang.  Variabel menunjukkan penyesuaian terbatas dan cenderung lambat merespons perubahan.
##      id variable
## 1 value    value

# example
df <- data.frame(
  time = 1:8,
  value = c(5, 5, 5, 5, 6, 6, 6, 7)
)

numspellr(df, lang = "english")

##        structure avg_spell median_spell max_spell stability_ratio
## 1 numeric_series       3.5          3.5         4           0.875
##   elasticity_index adjustment_frequency spell_concentration
## 1       0.05238095                 0.25           0.5102041
##      persistence_status
## 1 Very High Persistence
##                                                                                                                                                                                                                             interpretation
## 1 The series shows numeric persistence. Values remain unchanged for about 3.5 periods on average. Adjustments occur periodically. Changes tend to be sizable. From a policy perspective, this indicates strong rigidity and slow response.
##      id variable
## 1 value    value