diff --git a/docs/src/how_to_guides/timeseries.md b/docs/src/how_to_guides/timeseries.md
index 8c44478..c9e0da5 100644
--- a/docs/src/how_to_guides/timeseries.md
+++ b/docs/src/how_to_guides/timeseries.md
@@ -32,9 +32,23 @@ df.hour = Dates.hour.(df.Datetime)
Additionally, to simulate sudden changes caused by unforeseen events, such as blackouts or lockdowns, we deliberately reduce the electricity demand by 2GW from February 22nd onward.
``` julia
+df.Demand_updated = copy(df.Demand)
condition = df.Datetime .>= Date("2014-02-22")
-df[condition, :Demand] .= df[condition, :Demand] .- 2
+df[condition, :Demand_updated] .= df[condition, :Demand_updated] .- 2
```
+That is how the data looks like after our manipulation
+
+``` julia
+cutoff_point = 200
+plot(df[cutoff_point:split_index, [:Datetime]].Datetime, df[cutoff_point:split_index, :].Demand ,
+ label="training data", color=:green, xlabel = "Date" , ylabel="Electricity demand(GW)")
+plot!(df[split_index+1 : size(df,1), [:Datetime]].Datetime, df[split_index+1 : size(df,1), : ].Demand,
+ label="test data", color=:orange, xlabel = "Date" , ylabel="Electricity demand(GW)")
+plot!(df[split_index+1 : size(df,1), [:Datetime]].Datetime, df[split_index+1 : size(df,1), : ].Demand_updated, label="updated test data", color=:red, linewidth=1, framestyle=:box)
+plot!(legend=:outerbottom, legendcolumns=3)
+plot!(size=(850,400), left_margin = 5Plots.mm)
+```
+
### Lag features
@@ -54,10 +68,10 @@ df_dropped_missing
As usual, we split the data into train and test sets. We use the first 90% of the data for training and the remaining 10% for testing.
``` julia
-features_cols = DataFrames.select(df_dropped_missing, Not([:Datetime, :Demand]))
+features_cols = DataFrames.select(df_dropped_missing, Not([:Datetime, :Demand, :Demand_updated]))
X = Matrix(features_cols)
-y = Matrix(df_dropped_missing[:, [:Demand]])
-split_index = floor(Int, 0.9 * size(y , 1))
+y = Matrix(df_dropped_missing[:, [:Demand_updated]])
+split_index = floor(Int, 0.8 * size(y , 1))
println(split_index)
X_train = X[1:split_index, :]
y_train = y[1:split_index, :]
@@ -93,15 +107,30 @@ ub = [ maximum(tuple_data) for tuple_data in y_pred_interval]
y_pred = [mean(tuple_data) for tuple_data in y_pred_interval]
```
-
+``` julia
+cutoff_point = findfirst(df_dropped_missing.Datetime .== Date("2014-02-15"))
+plot(df_dropped_missing[cutoff_point:split_index, [:Datetime]].Datetime, y_train[cutoff_point:split_index] ,
+ label="train", color=:green , xlabel = "Date" , ylabel="Electricity demand(GW)", linewidth=1)
+plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime,
+ y_test, label="test", color=:red)
+plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime ,
+ y_pred, label ="prediction", color=:blue)
+plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime,
+ lb, fillrange = ub, fillalpha = 0.2, label = "prediction interval w/o EnbPI",
+ color=:lake, linewidth=0, framestyle=:box)
+plot!(legend=:outerbottom, legendcolumns=4, legendfontsize=6)
+plot!(size=(850,400), left_margin = 5Plots.mm)
+```
+
+
We can use `partial_fit` method in EnbPI implementation in ConformalPrediction in order to adjust prediction intervals to sudden change points on test sets that have not been seen by the model during training. In the below experiment, sample_size indicates the batch of new observations. You can decide if you want to update residuals by sample_size or update and remove first *n* residuals (shift_size = n). The latter will allow to remove early residuals that will not have a positive impact on the current observations.
-The chart below compares the results to the previous experiment without updating residuals:
+The figure below shows the new prediction intervals with updating residuals using EnbPI approach:
``` julia
-sample_size = 10
-shift_size = 10
+sample_size = 30
+shift_size = 100
last_index = size(X_test , 1)
lb_updated , ub_updated = ([], [])
for step in 1:sample_size:last_index
@@ -121,7 +150,22 @@ lb_updated = reduce(vcat, lb_updated)
ub_updated = reduce(vcat, ub_updated)
```
-
+``` julia
+plot(df_dropped_missing[cutoff_point:split_index, [:Datetime]].Datetime, y_train[cutoff_point:split_index] ,
+ label="train", color=:green , xlabel = "Date" , ylabel="Electricity demand(GW)", linewidth=1)
+plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime, y_test,
+ label="test", color=:red)
+plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime ,
+ y_pred, label ="prediction", color=:blue)
+plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime,
+ lb_updated, fillrange = ub_updated, fillalpha = 0.2, label = "EnbPI",
+ color=:lake, linewidth=0, framestyle=:box)
+ plot!(legend=:outerbottom, legendcolumns=4)
+ plot!(size=(850,400), left_margin = 5Plots.mm)
+```
+
+
+
## Results
diff --git a/docs/src/how_to_guides/timeseries.qmd b/docs/src/how_to_guides/timeseries.qmd
index 74f3bea..fbd0ae2 100644
--- a/docs/src/how_to_guides/timeseries.qmd
+++ b/docs/src/how_to_guides/timeseries.qmd
@@ -38,8 +38,21 @@ df.hour = Dates.hour.(df.Datetime)
Additionally, to simulate sudden changes caused by unforeseen events, such as blackouts or lockdowns, we deliberately reduce the electricity demand by 2GW from February 22nd onward.
```{julia}
+df.Demand_updated = copy(df.Demand)
condition = df.Datetime .>= Date("2014-02-22")
-df[condition, :Demand] .= df[condition, :Demand] .- 2
+df[condition, :Demand_updated] .= df[condition, :Demand_updated] .- 2
+```
+That is how the data looks like after our manipulation
+
+``` julia
+cutoff_point = 200
+plot(df[cutoff_point:split_index, [:Datetime]].Datetime, df[cutoff_point:split_index, :].Demand ,
+ label="training data", color=:green, xlabel = "Date" , ylabel="Electricity demand(GW)")
+plot!(df[split_index+1 : size(df,1), [:Datetime]].Datetime, df[split_index+1 : size(df,1), : ].Demand,
+ label="test data", color=:orange, xlabel = "Date" , ylabel="Electricity demand(GW)")
+plot!(df[split_index+1 : size(df,1), [:Datetime]].Datetime, df[split_index+1 : size(df,1), : ].Demand_updated, label="updated test data", color=:red, linewidth=1, framestyle=:box)
+plot!(legend=:outerbottom, legendcolumns=3)
+plot!(size=(850,400), left_margin = 5Plots.mm)
```
### Lag features
@@ -61,9 +74,9 @@ df_dropped_missing
As usual, we split the data into train and test sets. We use the first 90% of the data for training and the remaining 10% for testing.
```{julia}
-features_cols = DataFrames.select(df_dropped_missing, Not([:Datetime, :Demand]))
+features_cols = DataFrames.select(df_dropped_missing, Not([:Datetime, :Demand, :Demand_updated]))
X = Matrix(features_cols)
-y = Matrix(df_dropped_missing[:, [:Demand]])
+y = Matrix(df_dropped_missing[:, [:Demand_updated]])
split_index = floor(Int, 0.9 * size(y , 1))
println(split_index)
X_train = X[1:split_index, :]
@@ -100,17 +113,21 @@ ub = [ maximum(tuple_data) for tuple_data in y_pred_interval]
y_pred = [mean(tuple_data) for tuple_data in y_pred_interval]
```
-```{julia}
+``` julia
#| echo: false
#| output: true
-
cutoff_point = findfirst(df_dropped_missing.Datetime .== Date("2014-02-15"))
-p1 = plot(df_dropped_missing[cutoff_point:split_index, [:Datetime]].Datetime, y_train[cutoff_point:split_index] , label="train", color=:blue, legend=:bottomleft)
-plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime, y_test, label="test", color=:orange)
-plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime ,y_pred, label ="prediction", color=:green)
+plot(df_dropped_missing[cutoff_point:split_index, [:Datetime]].Datetime, y_train[cutoff_point:split_index] ,
+ label="train", color=:green , xlabel = "Date" , ylabel="Electricity demand(GW)", linewidth=1)
plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime,
- lb, fillrange = ub, fillalpha = 0.2, label = "PI without update of residuals", color=:green, linewidth=0)
-
+ y_test, label="test", color=:red)
+plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime ,
+ y_pred, label ="prediction", color=:blue)
+plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime,
+ lb, fillrange = ub, fillalpha = 0.2, label = "prediction interval w/o EnbPI",
+ color=:lake, linewidth=0, framestyle=:box)
+plot!(legend=:outerbottom, legendcolumns=4, legendfontsize=6)
+plot!(size=(850,400), left_margin = 5Plots.mm)
```
We can use `partial_fit` method in EnbPI implementation in ConformalPrediction in order to adjust prediction intervals to sudden change points on test sets that have not been seen by the model during training. In the below experiment, sample_size indicates the batch of new observations. You can decide if you want to update residuals by sample_size or update and remove first $n$ residuals (shift_size = n). The latter will allow to remove early residuals that will not have a positive impact on the current observations.
@@ -119,8 +136,8 @@ The chart below compares the results to the previous experiment without updating
```{julia}
-sample_size = 10
-shift_size = 10
+sample_size = 30
+shift_size = 100
last_index = size(X_test , 1)
lb_updated , ub_updated = ([], [])
for step in 1:sample_size:last_index
@@ -140,18 +157,20 @@ lb_updated = reduce(vcat, lb_updated)
ub_updated = reduce(vcat, ub_updated)
```
-```{julia}
+``` julia
#| echo: false
#| output: true
-
-p2 = plot(df_dropped_missing[cutoff_point:split_index, [:Datetime]].Datetime, y_train[cutoff_point:split_index] , label="train", color=:blue, legend=:bottomleft)
-plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime, y_test, label="test", color=:orange)
-plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime ,y_pred, label ="prediction", color=:green)
+plot(df_dropped_missing[cutoff_point:split_index, [:Datetime]].Datetime, y_train[cutoff_point:split_index] ,
+ label="train", color=:green , xlabel = "Date" , ylabel="Electricity demand(GW)", linewidth=1)
+plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime, y_test,
+ label="test", color=:red)
+plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime ,
+ y_pred, label ="prediction", color=:blue)
plot!(df_dropped_missing[split_index+1 : size(y,1), [:Datetime]].Datetime,
- lb_updated, fillrange = ub_updated, fillalpha = 0.2, label = "PI with adjusted residuals", color=:green, linewidth=0)
-
- plot(p1,p2, layout= (2,1))
-
+ lb_updated, fillrange = ub_updated, fillalpha = 0.2, label = "EnbPI",
+ color=:lake, linewidth=0, framestyle=:box)
+plot!(legend=:outerbottom, legendcolumns=4)
+plot!(size=(850,400), left_margin = 5Plots.mm)
```
## Results
diff --git a/docs/src/how_to_guides/timeseries_files/figure-commonmark/cell-10-output-1.svg b/docs/src/how_to_guides/timeseries_files/figure-commonmark/cell-10-output-1.svg
deleted file mode 100644
index b462309..0000000
--- a/docs/src/how_to_guides/timeseries_files/figure-commonmark/cell-10-output-1.svg
+++ /dev/null
@@ -1,54 +0,0 @@
-
-
diff --git a/docs/src/how_to_guides/timeseries_files/figure-commonmark/cell-12-output-1.svg b/docs/src/how_to_guides/timeseries_files/figure-commonmark/cell-12-output-1.svg
deleted file mode 100644
index 2b794af..0000000
--- a/docs/src/how_to_guides/timeseries_files/figure-commonmark/cell-12-output-1.svg
+++ /dev/null
@@ -1,94 +0,0 @@
-
-
diff --git a/docs/src/how_to_guides/timeseries_files/figure-commonmark/electricity_demand.svg b/docs/src/how_to_guides/timeseries_files/figure-commonmark/electricity_demand.svg
new file mode 100644
index 0000000..8eecf70
--- /dev/null
+++ b/docs/src/how_to_guides/timeseries_files/figure-commonmark/electricity_demand.svg
@@ -0,0 +1,48 @@
+
+
diff --git a/docs/src/how_to_guides/timeseries_files/figure-commonmark/electricity_demand_pred.svg b/docs/src/how_to_guides/timeseries_files/figure-commonmark/electricity_demand_pred.svg
new file mode 100644
index 0000000..f7d545a
--- /dev/null
+++ b/docs/src/how_to_guides/timeseries_files/figure-commonmark/electricity_demand_pred.svg
@@ -0,0 +1,56 @@
+
+
diff --git a/docs/src/how_to_guides/timeseries_files/figure-commonmark/electricity_demand_pred_enbpi.svg b/docs/src/how_to_guides/timeseries_files/figure-commonmark/electricity_demand_pred_enbpi.svg
new file mode 100644
index 0000000..8ff0545
--- /dev/null
+++ b/docs/src/how_to_guides/timeseries_files/figure-commonmark/electricity_demand_pred_enbpi.svg
@@ -0,0 +1,52 @@
+
+