Time Series Analysis

Chapter 4 · Seasonality and Forecasting

The slide of the lecture is on the left, the same figure on the right, computed in the browser from the same data. The ▶ button moves the parameter step by step.

Seasonality in economic data · Seasonal differencing and seasonal unit roots · SARIMA models · Case study: the airline model · Case study: seasonally unadjusted Romanian GDP · Seasonal adjustment in practice · Fourier terms and calendar effects · Multiple seasonality: electricity load · TBATS and Prophet · Forecast evaluation · Forecast combination

Seasonality in economic data

Seasonality in six Romanian series

Slide 7: Seasonality in six Romanian series
Slide 7
    Question for the room On 2010–2019, in which month is the mean monthly HICP inflation highest?
    Answer

    In January: 0.575% on average, followed by October, 0.487%; the lowest is in June, −0.368% (fresh food). The ranking depends on the period: on 2015–2019 October has 0.575% and May 0.457%. For GDP, the mean of quarter 4 is 57% above the mean of quarter 1 (27.8 and 43.6 bn EUR, 2000–2026); for tourism, August has 3.7 times as many nights as January (2015–2019).

    Seasonal plot and subseries plot

    Slide 9: Seasonal plot and subseries plot
    Slide 9
      Question for the room The ▶ button goes through the years 2012–2025. In which year is the ratio of August to January tourism nights largest?
      Answer

      In 2021: 5.3, against 3.4 in 2019 and 2.2 in 2020. In January 2021 restrictions still kept tourism low, and the summer recovered; in 2020 the summer was hit as well, and April 2020 had only 0.045 million nights. The shape of the year stays the same: August is the peak month in every year. For GDP, the four subseries keep the order Q1 < Q2 < Q3 < Q4 in all 21 years of 2005–2025.

      Seasonal differencing and seasonal unit roots

      The seasonal difference

      Slide 12: The seasonal difference
      Slide 12

      Worked example

      \[y_t=100\ln Y_t,\qquad \Delta_4y_t=(1-L^4)y_t=y_t-y_{t-4},\qquad \Delta\Delta_4y_t=(1-L)(1-L^4)y_t=y_t-y_{t-1}-y_{t-4}+y_{t-5}\]
          Question for the room In 2026Q2, Δy = 17.9 while Δ₄y = −1.96. Did GDP grow?
          Answer

          No: 17.9 is the seasonal jump from Q1 to Q2 (the GDP factors NSA/SCA are 0.78 in Q1 and 0.93 in Q2). The comparison with the same quarter removes the season: GDP in 2026Q2, 45.88 bn EUR, is 1.96 log points below 2025Q2, 46.79 bn EUR. ΔΔ₄y = −0.46 shows that the annual fall deepened, from −1.50 in 2026Q1 to −1.96.

          The roots of 1 - z^s: seasonal unit roots

          Slide 14: The roots of 1 - z^s: seasonal unit roots
          Slide 14
            Question for the room For daily data with a weekly pattern, s = 7. How many real roots does 1 − z⁷ have?
            Answer

            Only one, z = 1. The root z = −1 (period 2) exists only when s is even; for s = 7 the other six roots form three complex-conjugate pairs, with periods 7, 3.5 and 2.33 days. For s = 4: 1, −1 and ±i; for s = 12: 1, −1 and five pairs.

            Interpreting the seasonal roots

            Slide 15: Interpreting the seasonal roots
            Slide 15

            Worked example

            \[1-z^s=0\iff z_k=e^{2\pi ik/s}=\cos\frac{2\pi k}{s}+i\sin\frac{2\pi k}{s},\ \ k=0,1,\dots,s-1;\qquad (1-z_kz)(1-\bar z_kz)=1-2\cos\frac{2\pi k}{s}\,z+z^2\]
                Question for the room For monthly data (s = 12), what is the period of the root k = 1, and which real factor does it give with its conjugate?
                Answer

                Period 12/1 = 12 months, the annual cycle: z₁ = 0.866 + 0.500i, and the pair (k = 1, k = 11) gives the factor 1 − 1.732z + z². For s = 12 there are 12 roots: two real ones (1 and −1, period 2 months) and five complex-conjugate pairs, with periods 12, 6, 4, 3 and 2.4 months.

                Romanian GDP: regular and seasonal differences

                Slide 16: Romanian GDP: regular and seasonal differences
                Slide 16
                  Question for the room On the full sample, the ACF of ΔΔ₄ ln Y is −0.40 at lag 4. Is it still significant if the sample stops in 2019Q4, before the pandemic?
                  Answer

                  Yes: −0.34, outside the ±0.23 band (T = 75). The negative value at the seasonal lag does not come from 2020 but from a seasonal pattern that changes slowly: the signature of a seasonal MA. For Δ ln Y the ACF stays at 0.94 at lag 4 on the short sample as well; for Δ₄ ln Y the ACF at lag 1 is 0.68.

                  The HEGY test

                  Slide 18: The HEGY test
                  Slide 18

                  Worked example

                  \[\begin{gathered}y_{1t}=(1+L+L^2+L^3)y_t,\quad y_{2t}=-(1-L+L^2-L^3)y_t,\quad y_{3t}=-(1-L^2)y_t\\ \Delta_4y_t=\pi_1y_{1,t-1}+\pi_2y_{2,t-1}+\pi_3y_{3,t-2}+\pi_4y_{3,t-1}+\text{deterministic terms}+\textstyle\sum_{j=1}^{k}a_j\Delta_4y_{t-j}+\varepsilon_t\end{gathered}\]
                      Question for the room With 0 lags of Δ₄y instead of 1 (chosen by AIC), on 2000–2019, are all three nulls still not rejected?
                      Answer

                      No: with k = 0, t(π₂) = −3.34 < −2.77 and F(π₃, π₄) = 17.25 > 6.64, so the roots −1 and ±i are rejected; only the root 1 remains (t(π₁) = −1.46). With k = 1, chosen by AIC, no null is rejected (−1.97, −2.19, 6.12). The autocorrelation left in the residuals of the regression without lags changes the decision. The sample end matters too: up to 2025Q4, with k = 1, t(π₂) = −2.88 and F = 8.38 reject two nulls.

                      Seasonal unit-root tests on Romanian data

                      Slide 20: Seasonal unit-root tests on Romanian data
                      Slide 20
                        Question for the room For which series does the decision change when the sample end moves between 2012 and 2025?
                        Answer

                        Not for GDP: OCSB gives D = 1, Canova–Hansen D = 0 and KPSS d = 1 for every end. It changes for tourism nights (OCSB gives D = 1 for 2012 and 2020, and d moves between 0 and 1), for industrial production (D = 1 by OCSB for 2012 and 2014) and for the HICP (d = 1 for 2012, then d = 2). Canova–Hansen rejects stable seasonality for no series and no end.

                        SARIMA models

                        Worked example: the airline model

                        Slide 23: Worked example: the airline model
                        Slide 23

                        Worked example

                        \[(1-L)(1-L^{12})y_t=(1+\theta L)(1+\Theta L^{12})\varepsilon_t,\quad \gamma(0)=\sigma^2(1+\theta^2)(1+\Theta^2),\quad \rho_1=\frac{\theta}{1+\theta^2},\ \rho_{12}=\frac{\Theta}{1+\Theta^2},\ \rho_{11}=\rho_{13}=\rho_1\rho_{12}\]
                            Question for the room The airline model gives the same autocorrelation at lags 11 and 13. Do the airline passenger data show this?
                            Answer

                            Not exactly: the 1949–1960 estimates, θ̂ = −0.402 and Θ̂ = −0.557, imply ρ₁₁ = ρ₁₃ = 0.147, while the sample ACF of w_t is 0.064 at lag 11 and 0.152 at lag 13. Both values lie inside the band ±0.17, so the test does not separate them from the model value. At lags 1 and 12 the model implies −0.346 and −0.425, and the data give −0.341 and −0.387.

                            Theoretical ACF and PACF of three seasonal models

                            Slide 24: Theoretical ACF and PACF of three seasonal models
                            Slide 24
                              Question for the room With θ = −0.4 and Θ = −0.6, the satellites at lags 11 and 13 are 0.152. What happens to them if θ becomes +0.4?
                              Answer

                              They change sign: −0.152. The satellites are the product ρ₁ρ₁₂, and ρ₁ = θ/(1 + θ²) goes from −0.345 to 0.345, while ρ₁₂ = −0.441 stays. For SAR(1)₁₂ with Φ = 0.8 the ACF decays 0.8, 0.64, 0.512 at lags 12, 24, 36; for SMA(1)₁₂ with Θ = −0.6 the PACF is −0.242 at lag 24.

                              Forecasting with SARIMA

                              Slide 28: Forecasting with SARIMA
                              Slide 28

                              Worked example

                              \[\hat y_{T+h}=\hat y_{T+h-1}+\hat y_{T+h-12}-\hat y_{T+h-13}+\hat\theta\hat\varepsilon_{T+h-1}+\hat\Theta\hat\varepsilon_{T+h-12}+\hat\theta\hat\Theta\hat\varepsilon_{T+h-13},\qquad \hat\varepsilon_{T+j}=0\ (j\ge1)\]
                                  Question for the room From December 1958, how much do the MA terms add to the January 1959 forecast, and from which horizon do they vanish?
                                  Answer

                                  +0.0220 in logs, about 2.2%: the forecast is 348.6 thousand, against 341.0 without the MA terms; the actual value was 360. The MA terms use residuals up to lag 13 at most, so from h = 14 they are 0 and the forecast repeats the pattern on the line (February 1960: 354.6 against 391 actual).

                                  Case study: the airline model

                                  Identification of the airline model

                                  Slide 32: Identification of the airline model
                                  Slide 32
                                    Question for the room Without the regular difference, with Δ₁₂ ln y only, what does the ACF look like?
                                    Answer

                                    Positive autocorrelations that decay slowly: 0.71 at lag 1, 0.48 at lag 3, still 0.19 at lag 8. The trend is not removed, and the series stays non-stationary. With ΔΔ₁₂ ln y, non-zero values remain at lags 1 (−0.34) and 12 (−0.39); lags 3, 9, 23 and 32 also leave the ±0.17 band. The PACF is −0.341, −0.013 and −0.193 at lags 1, 2 and 3: the value at lag 2 is practically zero.

                                    Airline model: forecasts out of sample

                                    Slide 34: Airline model: forecasts out of sample
                                    Slide 34
                                      Question for the room With the origin in December 1958, the 95% interval for December 1960 has a width of 239. What width does it have if the origin is December 1954, 72 months ahead?
                                      Answer

                                      Much larger: 852 (the interval [201, 1053]), because the two unit roots make the forecast-error variance grow with the horizon. The actual value, 432, stays inside the interval from every origin. From 1958: forecast 388, interval [286, 526], MAPE 8.5% against 15.5% for the seasonal naive method.

                                      Interpreting the airline forecasts

                                      Slide 35: Interpreting the airline forecasts
                                      Slide 35
                                        Question for the room The slide gives Ljung–Box p = 0.37. Is this the test of the model estimated on 1949–1958?
                                        Answer

                                        No: for 1949–1958, p = 0.71; p = 0.37 is the test on the full sample, 1949–1960. The estimates depend on the sample length: on 1949–1952, Θ̂ = −0.904 with SE 3.39 (almost unidentified); on 1949–1958, Θ̂ = −0.541 with SE 0.105. With the origin in December 1952, the MAPE to December 1960 is 16.8% for the airline model and 38.2% for the seasonal naive method.

                                        Case study: seasonally unadjusted Romanian GDP

                                        SARIMA models for Romanian GDP

                                        Slide 36: SARIMA models for Romanian GDP
                                        Slide 36

                                        Worked example

                                        \[\mathrm{AIC}=-2\ln L+2k,\qquad \mathrm{AICc}=\mathrm{AIC}+\frac{2k(k+1)}{n-k-1},\qquad \mathrm{BIC}=-2\ln L+k\ln n\]
                                            Question for the room Which models fail the Ljung–Box test at 5% with m = 8, and does the BIC model pass with m = 12?
                                            Answer

                                            The three without a seasonal MA fail: (0,1,1)(1,1,0)₄ with p = 0.003, (1,1,0)(1,1,0)₄ with p = 0.004 and (0,1,1)(0,1,0)₄ with p = 0.005. The BIC model, (0,1,0)(0,1,1)₄, has p = 0.286 with m = 8 and p = 0.526 with m = 12. AICc differs by only 1.4 between this model and (1,1,1)(0,1,1)₄; BIC penalises the extra parameter more.

                                            Diagnostics of the GDP model

                                            Slide 38: Diagnostics of the GDP model
                                            Slide 38
                                              Question for the room SARIMA(0,1,1)(1,1,0)₄ replaces the seasonal MA by a seasonal AR. Are its residuals uncorrelated by the Ljung–Box test?
                                              Answer

                                              No: Q*(8) = 19.5 with 6 degrees of freedom, p = 0.003, and the AICc rises to 529.9, against 519.7. For SARIMA(0,1,0)(0,1,1)₄: Q*(8) = 8.55 with 7 degrees of freedom, p = 0.286, and Q*(12) has p = 0.526. In every model the residuals do not follow the Normal distribution: for the BIC model, Jarque–Bera 22.0 and the residual of −11.4% in 2020Q2.

                                              Forecasts of Romanian GDP

                                              Slide 40: Forecasts of Romanian GDP
                                              Slide 40
                                                Question for the room The ▶ button moves Θ from −0.95 to 0. What does the forecast year-on-year growth become when Θ = 0?
                                                Answer

                                                The same in every quarter: −1.96%, the year-on-year growth of 2026Q2. With Θ = 0 the model is Δ₄y_t = Δ₄y_{t−1} + ε_t, and the forecast repeats the last year-on-year growth. With Θ̂ = −0.577 the pattern gives weight 1 + Θ̂ = 0.42 to the most recent year: implied growth −0.81%, 0.53%, 0.83%, 0.55%, then 0.55%. With Θ = −0.95 the pattern is almost fixed, and the implied growth rises to 3.99%.

                                                Interpreting the GDP forecasts

                                                Slide 41: Interpreting the GDP forecasts
                                                Slide 41
                                                  Question for the room Θ = 0 means that last year's pattern repeats. How do the forecasts and the 95% interval for 2028Q2 change?
                                                  Answer

                                                  The forecasts fall a little: 52.5, 57.3, 37.6 and 45.0 bn EUR, against 53.1, 58.7, 38.7 and 46.1 with Θ = −0.577; the 2028Q2 interval becomes 32.4–60.0, against 37.6–57.2. With Θ = −0.95 (an almost fixed pattern) the interval is 37.3–66.1: the residual standard deviation rises when Θ moves away from the maximum-likelihood estimate.

                                                  Seasonal adjustment in practice

                                                  Unadjusted and adjusted series

                                                  Slide 46: Unadjusted and adjusted series
                                                  Slide 46
                                                    Question for the room In 2026Q1, unadjusted GDP falls by 42.1% (log points) from the previous quarter. How much did the SCA series change?
                                                    Answer

                                                    By −0.06%: practically not at all. The fall of the unadjusted series is the seasonal pattern, with an NSA/SCA factor of 0.786 in Q1 and 1.197 in 2025Q4. Year-on-year growth says the same in both series: −1.50% (NSA) and −1.14% (SCA). The annual totals coincide: 196.4 bn EUR in 2024.

                                                    Interpreting the adjusted series

                                                    Slide 47: Interpreting the adjusted series
                                                    Slide 47

                                                    Worked example

                                                    \[100(\ln Y_t-\ln Y_{t-1})\ \ \text{(NSA)},\qquad 100(\ln Y^{SCA}_t-\ln Y^{SCA}_{t-1})\ \ \text{(SCA)},\qquad \frac{Y_t}{Y^{SCA}_t}=\hat S_t\hat C_t\]
                                                        Question for the room Did the GDP seasonal factors change between 2010–2014 and 2020–2025?
                                                        Answer

                                                        Very little: Q1 0.77 and 0.79, Q2 0.93 and 0.93, Q3 1.09 and 1.08, Q4 1.20 and 1.20. The pattern moves slowly, as Θ̂ = −0.577 indicates. For retail trade, the January factor rises from 0.84 (2015–2019) to 0.86 (2020–2025), and the December one stays at 1.10.

                                                        Fourier terms and calendar effects

                                                        Seasonal dummies and Fourier terms

                                                        Slide 49: Seasonal dummies and Fourier terms
                                                        Slide 49

                                                        Worked example

                                                        \[\sum_{k=1}^{K}\Big[\alpha_k\sin\frac{2\pi kt}{m}+\beta_k\cos\frac{2\pi kt}{m}\Big],\qquad \text{amplitude}_k=\sqrt{\alpha_k^2+\beta_k^2}\]
                                                            Question for the room For daily load, on which day of 2025 is the annual Fourier component of the DHR highest, and how large is it?
                                                            Answer

                                                            On 1 February 2025: +0.64 GW (644 MW), with all four pairs; the lowest is on 13 May 2025, −0.60 GW. The first pair has an amplitude of 0.499 GW. For tourism, the K = 3 fit gives 0.765 for August, against the observed effect of 0.847.

                                                            Fourier approximation of a seasonal pattern

                                                            Slide 50: Fourier approximation of a seasonal pattern
                                                            Slide 50
                                                              Question for the room How many harmonics are needed for R² ≥ 0.98?
                                                              Answer

                                                              Three: K = 3, six coefficients, R² = 0.989. K = 1 gives 0.925, K = 2 gives 0.966. K = 5, with ten coefficients, gives 0.999. For K = 6, sin(2π6t/12) is zero in every month and is left out: with cos(2π6t/12), the 11 coefficients pass through all 12 points (R² = 1), like 11 dummy variables.

                                                              Interpreting the Fourier fits

                                                              Slide 51: Interpreting the Fourier fits
                                                              Slide 51
                                                                Question for the room Which K does AICc choose for the monthly tourism pattern?
                                                                Answer

                                                                K = 1 (AICc −41.5, against −40.1 at K = 2 and −32.0 at K = 3), although R² rises from 0.925 to 0.989 and the August peak needs K ≥ 3. With only 12 points the correction 2k(k + 1)/(12 − k − 1) grows fast: 28 at K = 3; at K = 5 and K = 6 AICc is no longer defined. K = 5 gives R² = 0.999; K = 6 (11 coefficients) passes through every point, R² = 1.

                                                                Calendar effects and Orthodox Easter

                                                                Slide 52: Calendar effects and Orthodox Easter
                                                                Slide 52

                                                                Worked example

                                                                \[E_t=\frac{1}{w}\,\#\{d\in[\text{Easter}-w,\ \text{Easter}-1]:\ d\ \text{in month}\ t\}\]
                                                                    Question for the room In 2027 Easter falls on 2 May. How is the regressor split with w = 10?
                                                                    Answer

                                                                    The window runs from 22 April to 1 May: 9 days in April (E = 0.9) and 1 day in May (E = 0.1). With β̂_E = 5.95% for food, the effect is +5.35% in April and +0.59% in May. In 2024 (5 May): 0.6 in April and 0.4 in May; in 2025 (20 April): 1 in April.

                                                                    Easter and Romanian retail trade

                                                                    Slide 54: Easter and Romanian retail trade
                                                                    Slide 54
                                                                      Question for the room With the window w = 3 (only the three days before Easter), does the estimated Easter effect on food sales rise or fall?
                                                                      Answer

                                                                      It falls to 1.6% (standard error 1.6, t = 1.0) and is no longer separated from noise. With w = 10 (lecture): 5.9% (2.3; t = 2.6); with w = 12: 6.1% (t = 2.9). Easter shopping spreads over several days; a regressor that is too narrow puts part of it in the wrong month. Working days stay around +0.6% for food and +1.0% for the total, for every w.

                                                                      Interpreting the Easter effect

                                                                      Slide 55: Interpreting the Easter effect
                                                                      Slide 55
                                                                        Question for the room Does the Easter effect on food retail depend on the window w?
                                                                        Answer

                                                                        Yes: with w = 5, β̂_E = 3.0% (SE 1.8, t = 1.7), not significant at 5%; with w = 10, 5.9% (SE 2.3, t = 2.6); with w = 20, 5.4% (SE 1.4, t = 3.8). For all retail the effect stays small (−0.1%, 0.8%, 1.5%), while working days matter at every w (about +1% per day).

                                                                        Multiple seasonality: electricity load

                                                                        Hourly load: the day and the week

                                                                        Slide 58: Hourly load: the day and the week
                                                                        Slide 58
                                                                          Question for the room For 2025, which autocorrelation is larger: at lag 24 (the same hour yesterday) or at lag 168 (the same hour a week ago)?
                                                                          Answer

                                                                          At lag 168: 0.817 against 0.811. The gap is larger in 2023 (0.841 and 0.800) and 2024 (0.846 and 0.804). At lag 84 (half a week) the ACF is only 0.22. Daily cycle: the minimum at 3:00 (5.3 GW), the maximum at 19:00 (7.3 GW).

                                                                          MSTL: one seasonal component per period

                                                                          Slide 60: MSTL: one seasonal component per period
                                                                          Slide 60
                                                                            Question for the room In the February 2025 window, the remainder has 2% of the variance around the trend. How much does it have in the window 7 April – 4 May 2025, which contains Easter?
                                                                            Answer

                                                                            13%: Easter (20 April) and 1 May follow neither the daily nor the weekly pattern, so they end up in the remainder. In February: 70% the daily component, 29% the weekly one. The share of the weekly component ranges from 20% (August) to 43% (June).

                                                                            Daily load and public holidays

                                                                            Slide 62: Daily load and public holidays
                                                                            Slide 62
                                                                              Question for the room Which is the day with the lowest load in 2023?
                                                                              Answer

                                                                              Pentecost Sunday, 4 June 2023: 4.37 GW, below Easter Sunday, 16 April 2023 (4.48 GW). In 2022, 2024 and 2025 the lowest day is Easter Sunday; over the whole sample, 20 April 2025 (3.79 GW, mean 6.18 GW). Saturday is 9% and Sunday 17% below Wednesday, and January is 25% above May.

                                                                              A dynamic harmonic regression for daily load

                                                                              Slide 64: A dynamic harmonic regression for daily load
                                                                              Slide 64

                                                                              Worked example

                                                                              \[y_t=\sum_{k=1}^{4}\Big[\alpha_k\sin\frac{2\pi kt}{365.25}+\beta_k\cos\frac{2\pi kt}{365.25}\Big]+\sum_j\delta_jD^{\mathrm{day}}_{jt}+\lambda_1H_t+\lambda_2E_t+\lambda_3X_t+u_t\]
                                                                                  Question for the room What does the deterministic part of the DHR give for Easter Monday, 21 April 2025, relative to a Sunday at the annual mean?
                                                                                  Answer

                                                                                  −0.145 GW: the Monday effect, +0.884, plus the annual component, −0.448, plus the Easter dummy, −0.581. An ordinary Monday at the same time of year would have about +0.44 GW. The level of the series comes from the ARIMA(1,1,1) errors, with AR 0.80 and MA −0.98.

                                                                                  TBATS and Prophet

                                                                                  Prophet components for daily load

                                                                                  Slide 68: Prophet components for daily load
                                                                                  Slide 68
                                                                                    Question for the room If Prophet is estimated without the list of holidays, what happens to the in-sample mean absolute error?
                                                                                    Answer

                                                                                    It rises from 197 to 226 MW. Without holidays, the Christmas and Easter days stay in the residuals and the yearly component is distorted: its maximum moves from 7 December to 24 January 2024. With the holidays, the effects are −1.24 GW (Christmas), −1.10 GW (New Year) and −1.04 GW (Easter).

                                                                                    Forecast evaluation

                                                                                    Time-series cross-validation

                                                                                    Slide 71: Time-series cross-validation
                                                                                    Slide 71
                                                                                      Question for the room With a step of 1 between origins and horizon h = 3, in how many forecast windows does the same date appear?
                                                                                      Answer

                                                                                      In three: consecutive windows overlap by h − 1 = 2 periods, so the errors are correlated up to lag h − 1, and the Diebold–Mariano test uses h − 1 autocovariances. In the design of the lecture the origins are 14 days apart and h = 14: the windows do not overlap, so the test has one value per window and h = 1 in the formula.

                                                                                      The Diebold–Mariano test

                                                                                      Slide 73: The Diebold–Mariano test
                                                                                      Slide 73
                                                                                        Question for the room Overall, DHR beats the seasonal naive method (HLN −4.42). At which horizons does the horizon-by-horizon test reject?
                                                                                        Answer

                                                                                        Only at h = 8 and h = 10–14 (|HLN| > 2.06); not at h = 1–7 and h = 9. A single day per origin is noisier than the 14-day average. DHR against Prophet: rejected at h = 1, 2 and 12–14.

                                                                                        Worked example: a Diebold–Mariano test

                                                                                        Slide 74: Worked example: a Diebold–Mariano test
                                                                                        Slide 74

                                                                                        Worked example

                                                                                        \[\bar d=\frac1n\sum_j d_j,\quad \hat\gamma_d(0)=\frac1n\sum_j(d_j-\bar d)^2,\quad \mathrm{DM}=\frac{\bar d}{\sqrt{\hat\gamma_d(0)/n}},\quad \mathrm{HLN}=\sqrt{\frac{n+1-2h+h(h-1)/n}{n}}\;\mathrm{DM}\]
                                                                                            Question for the room With the 26 real windows, DHR against the seasonal naive method: what are DM and HLN, and why do they differ only a little?
                                                                                            Answer

                                                                                            DM = −4.51, HLN = −4.42, p < 0.001; DHR has the smaller loss in 22 of the 26 windows (d̄ = −0.157 GW²). The HLN factor is √(25/26) = 0.981, so it matters little at n = 26; in the five-window example it is √(4/5) = 0.894, and the critical value t(4) = 2.78 instead of 1.96 changes the decision.

                                                                                            Forecasts from one origin, across Easter 2026

                                                                                            Slide 75: Forecasts from one origin, across Easter 2026
                                                                                            Slide 75
                                                                                              Question for the room In the Easter window (origin 6 April 2026), which model has the lowest MAE?
                                                                                              Answer

                                                                                              ETS, with 299 MW, followed by TBATS (309) and DHR (344); the seasonal naive method has 686, Prophet 616. ETS has no line in the chart of the slide, although it has the lowest error in this window. After Easter the load stays below all forecasts, and ETS and TBATS follow the level of the last days.

                                                                                              Cross-validation results for daily load

                                                                                              Slide 77: Cross-validation results for daily load
                                                                                              Slide 77
                                                                                                Question for the room In the 8 windows that contain a holiday, does DHR still have the lowest MAE?
                                                                                                Answer

                                                                                                No: TBATS 317 MW and DHR 318 MW are level, and the DM test does not separate them (HLN 0.83, p = 0.431). In the 18 windows without holidays, DHR has 258 MW, ahead of the combination (290). Over all 26 windows: DHR 276, the combination 305, Prophet 328 MW; ETS, SARIMA and TBATS are not separated from the seasonal naive method (p = 0.431, 0.231, 0.145).

                                                                                                Interpreting the cross-validation results

                                                                                                Slide 78: Interpreting the cross-validation results
                                                                                                Slide 78

                                                                                                Worked example

                                                                                                \[\mathrm{MASE}=\frac{\mathrm{MAE}}{\frac{1}{T-7}\sum_{t=8}^{T}|y_t-y_{t-7}|}\]
                                                                                                    Question for the room Without the 8 windows that contain holidays, is DHR still the best, and what happens in the holiday windows?
                                                                                                    Answer

                                                                                                    Without holidays (18 windows), DHR has MAE 258 MW and MASE 0.83, and the combination 0.94. In the 8 holiday windows no method has MASE below 1: TBATS 317 MW, DHR 318 MW, the combination 340 MW. Over horizons 8–14 not even DHR has MASE below 1 (1.02).

                                                                                                    A second test: Romanian GDP

                                                                                                    Slide 79: A second test: Romanian GDP
                                                                                                    Slide 79
                                                                                                      Question for the room Without 2019–2020, SARIMA has MAE 1.89 at h = 4, against 2.44 for ETS. Is the difference significant?
                                                                                                      Answer

                                                                                                      No: HLN = −1.11 (p = 0.275) at h = 4 and −1.47 (p = 0.148) at h = 1, with 43 origins. With all 51 origins, at h = 4 SARIMA even has MAE 2.70 against 2.84, and HLN = 0.50 (p = 0.620): the squared loss of SARIMA is larger in the pandemic years.

                                                                                                      Forecast combination

                                                                                                      Why combining forecasts works

                                                                                                      Slide 81: Why combining forecasts works
                                                                                                      Slide 81

                                                                                                      Worked example

                                                                                                      \[\mathrm{MSE}(w)=w^2\sigma_1^2+(1-w)^2\sigma_2^2+2w(1-w)\rho\sigma_1\sigma_2,\qquad w^*=\frac{\sigma_2^2-\rho\sigma_1\sigma_2}{\sigma_1^2+\sigma_2^2-2\rho\sigma_1\sigma_2}\]
                                                                                                          Question for the room With the real errors of DHR and Prophet, does the equal-weight average beat DHR alone?
                                                                                                          Answer

                                                                                                          No: σ₁ = 348 MW, σ₂ = 411 MW, ρ = 0.75; the variance of the equal combination is 126,036 MW², above σ₁² = 120,854. The best weight is w* = 0.82 (MSE 118,370), close to DHR alone. In MAE, the equal combination has 283 MW, DHR alone 276. Strongly correlated errors leave little room for a gain, as in the slide example (1.060 against 1).

                                                                                                          Combining DHR and Prophet

                                                                                                          Slide 82: Combining DHR and Prophet
                                                                                                          Slide 82
                                                                                                            Question for the room Combine ETS (MAE 381 MW) and Prophet (328 MW), whose errors have a correlation of 0.36. What MAE does the equal-weight average have?
                                                                                                            Answer

                                                                                                            295 MW: below both components. With weakly correlated errors the mistakes partly offset each other; the best weight in hindsight is w = 0.37 (MAE 286). For DHR and Prophet (ρ = 0.75): 283 MW with equal weights, against 276 for DHR alone; for ETS and TBATS (ρ = 0.98) the equal-weight average (370) is worse than TBATS alone (363).

                                                                                                            Interpreting the combination

                                                                                                            Slide 83: Interpreting the combination
                                                                                                            Slide 83
                                                                                                              Question for the room Do the weights estimated on the earlier windows beat equal weights for the five models?
                                                                                                              Answer

                                                                                                              A little: MAE 299 MW, against 305 with equal weights; weights chosen with hindsight give 296, and DHR alone 276. For DHR and Prophet: 283 with equal weights, 281 from the earlier windows, 274 with the best weight in hindsight (w = 0.86). The gain of estimated weights is small, as the forecast combination puzzle says.

                                                                                                              Romanian GDP: Eurostat, chain-linked volumes (2010 prices), not adjusted (NSA) and seasonally and calendar adjusted (SCA), in the version used in the lecture. Retail trade, industrial production and tourism nights: Eurostat, up to July 2026; HICP: Eurostat, up to August 2026. Electricity load: ENTSO-E, “monthly hourly load values”, Romania, 1 January 2022 – 30 June 2026, Romanian winter time. Airline passengers: the AirPassengers data set (Box and Jenkins), 1949–1960. The cross-validation forecasts are those of the lecture; variants that do not appear in the lecture are extra runs of the chapter code, on the same data.