The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation

Mathematical background

In the following formulas, X_i is the predicted i^th value, and the Y_i element is the actual i^th value. The regression method predicts the X_i element for the corresponding Y_i element of the ground truth dataset. Define two constants: the mean of the true values

(1) $$\overline{Y}={\displaystyle \frac{1}{m}\sum _{i=1}^m{Y}_i}$$

and the mean total sum of squares

(2) $$\mathrm{M}\mathrm{S}\mathrm{T}={\displaystyle \frac{1}{m}\sum _{i=1}^m(Y_i-\overline{Y}{)}^2}$$

Coefficient of determination (R² or R-squared)

(3) $${\mathrm{R}}^2=1-{\displaystyle \frac{\sum _{i=1}^m{(X_i-Y_i)}^2}{\sum _{i=1}^m{(\overline{Y}-Y_i)}^2}}$$

(worst value = −∞; best value = +1)

The coefficient of determination (Wright, 1921) can be interpreted as the proportion of the variance in the dependent variable that is predictable from the independent variables.

Mean square error (MSE)

(4) $$\mathrm{M}\mathrm{S}\mathrm{E}={\displaystyle \frac{1}{m}\sum _{i=1}^m(X_i-Y_i{)}^2}$$

(best value = 0; worst value = +∞)

MSE can be used if there are outliers that need to be detected. In fact, MSE is great for attributing larger weights to such points, thanks to the L₂ norm: clearly, if the model eventually outputs a single very bad prediction, the squaring part of the function magnifies the error.

Since $\mathrm{R}$² $=1-{\displaystyle \frac{\mathrm{M}\mathrm{S}\mathrm{E}}{\mathrm{M}\mathrm{S}\mathrm{T}}}$ and since MST is fixed for the data at hand, R² is monotonically related to MSE (a negative monotonic relationship), which implies that an ordering of regression models based on R² will be identical (although in reverse order) to an ordering of models based on MSE or RMSE.

Root mean square error (RMSE)

(5) $$\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E}=\sqrt{{\displaystyle \frac{1}{m}\sum _{i=1}^m{(X_i-Y_i)}^2}}$$

(best value = 0; worst value = +∞)

The two quantities MSE and RMSE are monotonically related (through the square root). An ordering of regression models based on MSE will be identical to an ordering of models based on RMSE.

Mean absolute error (MAE)

(6) $$\mathrm{M}\mathrm{A}\mathrm{E}={\displaystyle \frac{1}{m}\sum _{i=1}^m|X_i-Y_i|}$$

(best value = 0; worst value = +∞)

MAE can be used if outliers represent corrupted parts of the data. In fact, MAE is not penalizing too much the training outliers (the L₁ norm somehow smooths out all the errors of possible outliers), thus providing a generic and bounded performance measure for the model. On the other hand, if the test set also has many outliers, the model performance will be mediocre.

Mean absolute percentage error (MAPE)

(7) $$\mathrm{M}\mathrm{A}\mathrm{P}\mathrm{E}={\displaystyle \frac{1}{m}\sum _{i=1}^m|{\displaystyle \frac{Y_i-X_i}{Y_i}|}}$$

(best value = 0; worst value = +∞)

MAPE is another performance metric for regression models, having a very intuitive interpretation in terms of relative error: due to its definition, its use is recommended in tasks where it is more important being sensitive to relative variations than to absolute variations (De Myttenaere et al., 2016). However, its has a number of drawbacks, too, the most critical ones being the restriction of its use to strictly positive data by definition and being biased towards low forecasts, which makes it unsuitable for predictive models where large errors are expected (Armstrong & Collopy, 1992).

Symmetric mean absolute percentage error (SMAPE)

(8) $$\mathrm{S}\mathrm{M}\mathrm{A}\mathrm{P}\mathrm{E}={\displaystyle \frac{100\mathrm{\%}}{m}\sum _{i=1}^m{\displaystyle \frac{|X_i-Y_i|}{(|X_i|+|Y_i|)/2}}}$$

(best value = 0; worst value = 2)

Initially defined by Armstrong (1985), and then refined in its current version by Flores (1986) and Makridakis (1993), SMAPE was proposed to amend the drawbacks of the MAPE metric. However, there is little consensus on a definitive formula for SMAPE, and different authors keep using slightly different versions (Hyndman, 2014). The original SMAPE formula defines the maximum value as 200%, which is computationally equivalent to 2. In this manuscript, we are going to use the first value for formal passages, and the second value for numeric calculations.

Informativeness

The rates RMSE, MAE, MSE and SMAPE have value 0 if the linear regression model fits the data perfectly, and positive value if the fit is less than perfect. Furthermore, the coefficient of determination has value 1 if the linear regression model fits the data perfectly (that means if MSE = 0), value 0 if MSE = MST, and negative value if the mean squared error, MSE, is greater than mean total sum of squares, MST.

Even without digging into the mathematical properties of the aforementioned statistical rates, it is clear that it is difficult to interpret sole values of MSE, RMSE, MAE and MAPE, since they have +∞ as upper bound. An MSE = 0.7, for example, does not say much about the overall quality of a regression model : the value could mean both an excellent regression model and a poor regression model . We cannot know it unless the maximum MSE value for the regression task is provided or unless the distribution of all the ground truth values is known. The same concept is valid for the other rates having +∞ as upper bound, such as RMSE, MAE and MAPE.

The only two regression scores that have strict real values are the non-negative R-squared and SMAPE. R-squared can have negative values, which mean that the regression performed poorly. R-squared can have value 0 when the regression model explains none of the variability of the response data around its mean (Minitab Blog Editor, 2013).

The positive values of the coefficient of determination range in the [0, 1] interval, with 1 meaning perfect prediction. On the other side, the values of SMAPE range in the [0, 2], with 0 meaning perfect prediction and 2 meaning worst prediction possible.

This is the main advantage of the coefficient of determination and SMAPE over RMSE, MSE, MAE, and MAPE: values like R² = 0.8 and SMAPE = 0.1, for example, clearly indicate a very good regression model performance, regardless of the ranges of the ground truth values and their distributions. A value of RMSE, MSE, MAE, or MAPE equal to 0.7, instead, fails to inform us about the quality of the regression performed.

This property of R-squared and SMAPE can be useful in particular when one needs to compare the predictive performance of a regression on two different datasets having different value scales. For example, suppose we have a mental health study describing a predictive model where the outcome is a depression scale ranging from 0 to 100, and another study using a different depression scale, ranging from 0 to 10 (Reeves, 2021). Using R-squared or SMAPE we could compare the predictive performance of the two studies without making additional transformations. The same comparison would be impossible with RMSE, MSE, MAE, or MAPE.

Given the better robustness of R-squared and SMAPE over the other four rates, we focus the rest of this article on the comparison between these two statistics.

R-squared and SMAPE

R-squared

The coefficient of determination can take values in the range (−∞, 1] according to the mutual relation between the ground truth and the prediction model. Hereafter we report a brief overview of the principal cases.

R² ≥ 0: With linear regression with no constraints, R² is non-negative and corresponds to the square of the multiple correlation coefficient.

R² = 0: The fitted line (or hyperplane) is horizontal. With two numerical variables this is the case if the variables are independent, that is, are uncorrelated. Since $\mathrm{R}$² $=1-{\displaystyle \frac{\mathrm{M}\mathrm{S}\mathrm{E}}{\mathrm{M}\mathrm{S}\mathrm{T}}}$, the relation R² = 0 is equivalent to MSE = MST, or, equivalently, to:

(9) $$\sum _{i=1}^m(Y_i-\overline{Y}{)}^2=\sum _{i=1}^m(Y_i-X_i{)}^2$$

Now, Eq. 9 has the obvious solution X_i = $\overline{Y}$ for 1 ≤ i ≤ m, but, being just one quadratic equation with m unknowns X_i, it has infinite solutions, where X_i = $\overline{Y}$ ± ɛ_i for a small ɛ_i, as shown in the following example:

• {Y_i 1 ≤ i ≤ 10} = {90.317571, 40.336481, 5.619065,44.529437, 71.192687, 32.036909, 6.977097, 66.425010, 95.971166, 5.756337}

• Ȳ= 45.91618

• {X_i 1 ≤ i ≤ 10} = {45.02545, 43.75556, 41.18064, 42.09511, 44.85773, 44.09390, 41.58419, 43.25487, 44.27568, 49.75250}

• MSE = MST = 1051.511

• R² ≈ 10⁻⁸ .

R² < 0: This case is only possible with linear regression when either the intercept or the slope are constrained so that the "best-fit" line (given the constraint) fits worse than a horizontal line, for instance if the regression line (hyperplane) does not follow the data (CrossValidated, 2011b). With nonlinear regression, the R-squared can be negative whenever the best-fit model (given the chosen equation, and its constraints, if any) fits the data worse than a horizontal line. Finally, negative R² might also occur when omitting a constant from the equation, that is, forcing the regression line to go through the point (0,0).

A final note. The behavior of the coefficient of determination is rather independent from the linearity of the regression fitting model: R² can be very low even for completely linear model, and vice versa, a high R² can occur even when the model is noticeably non-linear. In particular, a good global R² can be split in several local models with low R² (CrossValidated, 2011a).

SMAPE

By definition, SMAPE values range between 0% and 200%, where the following holds in the two extreme cases:

SMAPE = 0: The best case occurs when SMAPE vanishes, that is when

$${\displaystyle \frac{100\mathrm{\%}}{m}\sum _{i=1}^m{\displaystyle \frac{|X_i-Y_i|}{(|X_i|+|Y_i|)/2}=0}}$$

equivalent to

$$\sum _{i=1}^m{\displaystyle \frac{|X_i-Y_i|}{(|X_i|+|Y_i|)/2}=0}$$

and, since the m components are all positive, equivalent to

$${\displaystyle \frac{|X_i-Y_i|}{|X_i|+|Y_i|}=0\mathrm{\forall }1\le i\le m}$$

and thus X_i = Y_i, that is, perfect regression.

SMAPE = 2: The worst case SMAPE = 200% occurs instead when

$${\displaystyle \frac{100\mathrm{\%}}{m}\sum _{i=1}^m{\displaystyle \frac{|X_i-Y_i|}{(|X_i|+|Y_i|)/2}=2}}$$

equivalent to

$$\sum _{i=1}^m{\displaystyle \frac{|X_i-Y_i|}{|X_i|+|Y_i|}=m}$$

By the triangle inequality |a + c| ≤ |a| + | c| computed for b = −c, we have that |a – b| ≤ |a| + |b| and thus ${\displaystyle \frac{|a-b|}{|a|+|b|}\le 1}$. This yields that SMAPE = 2 if ${\displaystyle \frac{|X_i-Y_i|}{|X_i|+|Y_i|}=1}$ for all i = 1,…,m. Thus we reduced to compute when $\xi (a,b)={\displaystyle \frac{|a-b|}{|a|+|b|}=1}$: we analyse now all possible cases, also considering the symmetry of the relation with respect to a and b, ξ(a,b) = ξ(b,a).

If a = 0, $\xi (0,b)={\displaystyle \frac{|0-b|}{|0|+|b|}=1}$ if b = 0.

Now suppose that a,b > 0: ξ(a,a) = 0, so we can suppose a > b, thus a = b + ε, with a,b,ε > 0. Then $\xi (a,b)=\xi (b+\epsilon ,\epsilon )={\displaystyle \frac{\epsilon }{2b+\epsilon }<1}$. Same happens when a,b < 0: thus, if ground truth points and the prediction points have the same sign, SMAPE will never reach its maximum value.

Finally, suppose that a and b have opposite sign, for instance a > 0 and b < 0. Then b = −c, for c > 0 and thus $\xi (a,b)=\xi (a,-c)={\displaystyle \frac{|a+c|}{|a|+|c|}={\displaystyle \frac{a+c}{a+c}=1}}$.

Summarising, SMAPE reaches its worst value 200% if

• X_i = 0 and Y_i = 0 for all i = 1,…,m

• X_i = 0 and Y_i = 0 for all i = 1,…,m

• X_i· Y_i<0 for all i = 1,…,m, that is, ground truth and prediction always have opposite sign, regardless of their values.

For instance, if the ground truth points are (1, −2, 3, −4, 5, −6, 7, −8, 9, −10) , any prediction vector with all opposite signs (for example, (−307.18, 636.16, −469.99, 671.53, −180.55, 838.23, −979.18 , 455.16, −8.32, 366.80) ) will result in a SMAPE metric reaching 200%.

Explained the extreme cases of R-squared and SMAPE, in the next section we illustrate some significant, informative use cases where these two rates generate discordant outcomes.

Use cases

We list hereafter a number of example use cases where the coefficient of determination and SMAPE produce divergent outcomes, showing that R² is more robust and reliable than SMAPE, especially on bad poor quality regressions. To simplify comparison between the two measures, define the complementary normalized SMAPE as:

(10) $$\mathrm{c}\mathrm{n}\mathrm{S}\mathrm{M}\mathrm{A}\mathrm{P}\mathrm{E}=1-{\displaystyle \frac{\mathrm{S}\mathrm{M}\mathrm{A}\mathrm{P}\mathrm{E}}{200\mathrm{\%}}}$$

(worst value = 0; best value = 1)

UC1 use case

Consider the ground truth set $REAL=\{r_i=(i,i)\in {\mathbb{R}}^2,i\in \mathbb{N},1\le i\le 100\}$ collecting 100 points with positive integer coordinates on the straight line y = x. Define then the set PRED_j = {p_i} as

(11) $$p_i=(\begin{array}{cc}{r}_i\hfill & \mathrm{i}\mathrm{f}i\text{⧸}\equiv 1(mod5)\hfill \\ r_{5k+1}\hfill & \mathrm{f}\mathrm{o}\mathrm{r}k\ge j\hfill \\ 0\hfill & \mathrm{f}\mathrm{o}\mathrm{r}i=5k+1,0\le k<j\hfill \end{array}$$

so that REAL and PRED_j coincides apart from the first j points 1, 6, 11,… congruent to 1 modulo 5 that are set to 0. Then, for each 5 ≤ j ≤ 20, compute R² and cnSMAPE (Table 1).

Both measures decrease with the increasing number of non-matching points p_{5k + 1} = 0, but cnSMAPE decreases linearly, while R² goes down much faster, better showing the growing unreliability of the predicted regression. At the end of the process, j = 20 points out of 100 are wrong, but still cnSMAPE is as high as 0.80, while R² is 0.236, correctly declaring PRED₂₀ a very weak prediction set.

UC2 use case

In a second example, consider again the same REAL dataset and define the three predicting sets

$$\begin{array}{cc}\mathrm{P}\mathrm{R}\mathrm{E}{\mathrm{D}}_{}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{r}\mathrm{t}\hfill & =\{p_i^s:1\le i\le 100\}\hfill \\ p_i^s\hfill & =\{\begin{array}{cc}{r}_i\hfill & \mathrm{f}\mathrm{o}\mathrm{r}i\ge 10\hfill \\ 0\hfill & \mathrm{f}\mathrm{o}\mathrm{r}i<10\hfill \end{array}\hfill \\ \mathrm{P}\mathrm{R}\mathrm{E}{\mathrm{D}}_{}\mathrm{m}\mathrm{i}\mathrm{d}\mathrm{d}\mathrm{l}\mathrm{e}\hfill & =\{p_i^m:1\le i\le 100\}\hfill \\ p_i^m\hfill & =\{\begin{array}{cc}{r}_i\hfill & \mathrm{f}\mathrm{o}\mathrm{r}i\le 50andi\ge 61\hfill \\ 0\hfill & \mathrm{f}\mathrm{o}\mathrm{r}51\le i\le 60\hfill \end{array}\hfill \\ \mathrm{P}\mathrm{R}\mathrm{E}{\mathrm{D}}_{}\mathrm{e}\mathrm{n}\mathrm{d}\hfill & =\{p_i^e:1\le i\le 100\}\hfill \\ p_i^e\hfill & =\{\begin{array}{cc}{r}_i\hfill & \mathrm{f}\mathrm{o}\mathrm{r}i\le 90\hfill \\ 0\hfill & \mathrm{f}\mathrm{o}\mathrm{r}i\ge 91\hfill \end{array}\hfill \end{array}$$

In all the three cases start, middle, end the predicting set coincides with REAL up to 10 points that are set to zero, at the beginning, in the middle and at the end of the prediction, respectively. Interestingly, cnSMAPE is 0.9 in all the three cases, showing that SMAPE is sensible only to the number of non-matching points, and not to the magnitude of the predicting error. R² instead correctly decreases when the zeroed sequence of points is further away in the prediction and thus farthest away from the actual values: R² is 0.995 for PRED_start, 0.6293 for PRED_middle and −0.0955 for PRED_end.

UC3 use case

Consider now the as the ground truth the line y = x, and sample the set T including twenty positive integer points T = {t_i = (x_i,y^T_i) = (i,i) 1 ≤ i ≤ 20} on the line. Define REAL = {r_i = (x_i,y^R_i) = (i,i + N(i)) 1 ≤ i ≤ 20} as the same points of T with a small amount of noise N(i) on the y axes, so that r_i are close but not lying on the y = x straight line. Consider now two predicting regression models:

• The set PRED_c = T representing the correct model;

• The set PRED_w representing the (wrong) model with points defined as p^w_i = f(x_i), for f the 10-th degree polynomial exactly passing through the points r_i for 1 ≤ i ≤ 10.

Clearly, p^w_i coincides with r_i for 1 ≤ i ≤ 10, but ||p^w_i − r_i|| becomes very large for i ≥ 11. On the other hand t_i ≠ r_i for all i’s, but ||t_i − r_i|| is always very small. Compute now the two measures R² and cnSMAPE on the first N points i = 1, …, N for 2 ≤ N ≤ 20 of the two different regression models c and w with respect to the ground truth set REAL (Table 2).

For the correct regression model, both measures are correctly showing good results. For the wrong model, both measures are optimal for the first 10 points, where the prediction exactly matches the actual values; after that, R² rapidly decreases supporting the inconsistency of the model, while cnSMAPE is not affected that much, arriving for N = 20 to a value 1/2 as a minimum, even if the model is clearly very bad in prediction.

UC4 use case

Consider the following example: the seven actual values are (1, 1, 1, 1, 1, 2, 3) , and the predicted values are (1, 1, 1, 1, 1, 1, 1) . From the predicted values, it is clear that the regression method worked very poorly: it predicted 1 for all the seven values.

If we compute the coefficient of determination and SMAPE here, we obtain R-squared = −0.346 and SMAPE = 0.238. The coefficient of determination illustrates that something is completely off, by having a negative value. On the contrary, SMAPE has a very good score, that corresponds to 88.1% correctness in the cnSMAPE scale.

In this use case, if a inexperienced practitioner decided to check only the value of SMAPE to evaluate her/his regression, she/he would be misled and would wrongly believe that the regression went 88.1% correct. If, instead, the practitioner decided to verify the value of R-squared, she/he would be alerted about the poor quality of the regression. As we saw earlier, the regression method predicted 1 for all the seven ground truth elements, so it clearly performed poorly.

UC5 use case

Let us consider now a vector of 5 integer elements having values (1, 2, 3, 4, 5) , and a regression prediction made by the variables (a, b, c, d, e) . Each of these variables can assume all the integer values between 1 and 5, included. We compute the coefficient of determination and cnSMAPE for each of the predictions with respect to the actual values. To compare the values of the coefficient of determination and cnSMAPE in the same range, we consider only the cases when R-squared is greater or equal to zero, and we call it non-negative R-squared. We reported the results in Fig. 1.

As clearly observable in the plot Fig. 1, there are a number of points where cnSMAPE has a high value (between 0.6 and 1) but R-squared had value 0: in these cases, the coefficient of determination and cnSMAPE give discordant outcomes. One of these cases, for example, is the regression where the predicted values have values (1, 2, 3, 5, 2) , R² = 0, and cnSMAPE = 0.89.

In this example, cnSMAPE has a very high value, meaning that the prediction is 89% correct, while R² is equal to zero. The regression correctly predicts the first three points (1, 2, 3) , but fails to classify the forth element (4 is wrongly predicted as 5), and the fifth element (5 is mistakenly labeled as 2). The coefficient of determination assigns a bad outcome to this regression because it fails to correctly classify the only members of the 4 and 5 classes. Diversely, SMAPE assigns a good outcome to this prediction because the variance between the actual values and the predicted values is low, in proportion to the overall mean of the values.

Faced with this situation, we consider the outcome of the coefficient of determination more reliable and trustworthy: similarly to the Matthews correlation coefficient (MCC) (Matthews, 1975) in binary classification (Chicco & Jurman, 2020; Chicco, Tötsch & Jurman, 2021; Tötsch & Hoffmann, 2021; Chicco, Starovoitov & Jurman, 2021; Chicco, Warrens & Jurman, 2021), R-squared generates a high score only if the regression is able to correctly classify most of the elements of each class. In this example, the regression fails to classify all the elements of the 4 class and of the 5 class, so we believe a good metric would communicate this key-message.

Medical scenarios

To further investigate the behavior of R-squared, MAE, MAPE, MSE, RMSE and SMAPE, we employed these rates to a regression analysis applied to two real biomedical applications.

Hepatitis dataset

We trained and applied several machine learning regression methods on the Lichtinghagen dataset (Lichtinghagen et al., 2013; Hoffmann et al., 2018), which consists of electronic health records of 615 individuals including healthy controls and patients diagnosed with cirrhosis, fibrosis, and hepatitis. This dataset has 13 features, including a numerical variable stating the diagnosis of the patient, and is publicly available in the University of California Irvine Machine Learning Repository (2020). There are 540 healthy controls (87.8%) and 75 patients diagnosed with hepatitis C (12.2%). Among the 75 patients diagnosed with hepatitis C, there are: 24 with only hepatitis C (3.9%); 21 with hepatitis C and liver fibrosis (3.41%); and 30 with hepatitis C, liver fibrosis, and cirrhosis (4.88%).

Obesity dataset

To further verify the effect of the regression rates, we applied the data mining methods to another medical dataset made of electronic health records of young patients with obesity (Palechor & De-La-Hoz-Manotas, 2019, De-La-Hoz-Correa et al., 2019). This dataset is publicly available in the University of California Irvine Machine Learning Repository (2019) too, and contains data of 2,111 individuals, with 17 variables for each of them. A variable called NObeyesdad indicates the obesity level of each subject, and can be employed as a regression target. In this dataset, there are 272 children with insufficient weight (12.88%), 287 children with normal weight (13.6%), 351 children with obesity type I (16.63%), 297 children with obesity type II (14.07%), 324 children with obesity type III (15.35%), 290 children with overweight level I (13.74%), and 290 children with overweight level II (13.74%). The original curators synthetically generated part of this dataset (Palechor & De-La-Hoz-Manotas, 2019, De-La-Hoz-Correa et al., 2019).

Methods

For the regression analysis, we employed the same machine learning methods two of us authors used in a previous analysis (Chicco & Jurman, 2021): Linear Regression (Montgomery, Peck & Vining, 2021), Decision Trees (Rokach & Maimon, 2005), and Random Forests (Breiman, 2001), all implemented and executed in the R programming language (Ihaka & Gentleman, 1996). For each method execution, we first shuffled the patients data, and then we randomly selected 80% of the data elements for the training set and used the remaining 20% for the test set. We trained each method model on the training set, applied the trained model to the test set, and saved the regression results measured through R-squared, MAE, MAPE, MSE, RMSE, and SMAPE. For the hepatitis dataset , we imputed the missing data with the Predictive Mean Matching (PMM) approach through the Multiple Imputation by Chained Equations (MICE) method (Buuren & Groothuis-Oudshoorn, 2010). We ran 100 executions and reported the results means and the rankings based on the different rates in Table 3 (hepatitis dataset) and in Table 4 (obesity dataset).

Hepatitis dataset results: different rate, different ranking

We measured the results obtained by these regression models on the Lichtinghagen hepatitis dataset with all the rates analyzed in our study: R², MAE, MAPE, RMSE, MSE and SMAPE (lower part of Table 3).

These rates generate 3 different rankings. R², MSE and RMSE share the same ranking (Random Forests, Linear Regression and Decision Tree). SMAPE and MAPE share the same ranking (Decision Tree, Random Forests and Linear Regression). MAE has its own ranking (Random Forests, Decision Tree and Linear Regression).

It is also interesting to notice that these six rates select different methods as top performing method. R², MAE, MSE and RMSE indicate Random Forests as top performing regression model, while SMAPE and MAPE select Decision Tree for the first position in their rankings. The position of Linear Regression changes, too: on the second rank for R², MSE and RMSE, while on the last rank for MAE, SMAPE and MAPE.

By comparing all these different standings, a machine learning practitioner could wonder what is the most suitable rate to choose, to understand how the regression experiments actually went and which method outperformed the others. As explained earlier, we suggest the readers to focus on the ranking generated by the coefficient of determination, because it is the only metric that considers the distribution of all the ground truth values, and generates a high score only if the regression correctly predict most of the values of each ground truth category. Additionally, the fact that the ranking indicated by R-squared (Random Forests, Linear Regression and Decision Tree) was the same standing generated by 3 rates out of 6 suggests that it is the most informative one (Table 3).

Hepatitis dataset results: R² provides the most informative outcome

Another interesting aspect of these results on the hepatitis dataset regards the comparison between coefficient of determination and SMAPE (Table 3). We do not compare the standing of R-squared with MAE, MSE, RMSE, and MAPE because these four rates can have infinite positive values and, as mentioned earlier, this aspect makes it impossible to detect the quality of a regression from a single score of these rates.

R-squared indicates a very good result for Random Forests (R² = 0.756), and good results for Linear Regression (R² = 0.535) and Decision Tree (R² = 0.423). On the contrary, SMAPE generates an excellent result for Decision Tree (SMAPE = 0.073), meaning almost perfect prediction, and poor results for Random Forests (SMAPE = 1.808) and Linear Regression (SMAPE = 1.840), very close to the upper bound (SMAPE = 2) representing the worst possible regression.

These values mean that the coefficient of determination and SMAPE generate discordant outcomes for these two methods: for R-squared, Random Forests made a very good regression and Decision Tree made a good one; for SMAPE, instead, Random Forests made a catastrophic regression and Decision Tree made an almost perfect one. At this point, a practitioner could wonder which algorithm between Random Forests and Decision Trees made the better regression. Checking the standings of the other rates, we clearly see that Random Forests resulted being the top model for 4 rates out of 6, while Decision Tree resulted being the worst model for 3 rates out of 6. This information confirms that the ranking of R-squared is more reliable than the one of SMAPE (Table 3).

Obesity dataset results: agreement between rankings, except for SMAPE

Differently from the rankings generated on the hepatitis dataset, the rankings produced on the obesity dataset are more concordant (Table 4). Actually, the ranking of the coefficient of determination, MSE, RMSE, MAE and MAPE are identical: Random Forests on the first position, Decision Tree on the second position, and Linear Regression on the third and last position. All the rates’ rankings indicate Random Forests as the top performing method.

The only significant difference can be found in the SMAPE standing: differently from the other rankings that all put Decision Tree as second best regressor and Linear Regression as worst regressor, the SMAPE standing indicates Linear Regression as runner-up and Decision Tree on the last position. SMAPE, in fact, swaps the positions of these two methods, compared to R-squared and the other rates: SMAPE says Linear Regression outperformed Decision Tree, while the other rates say that Decision Tree outperformed Linear Regression.

Since five out of six rankings confirm that Decision Tree generated better results than Linear Regression, and only one of six say vice versa, we believe that is clear that the ranking indicated by the coefficient of determination is more informative and trustworthy than the ranking generated by SMAPE.