MUDITA: A New Approach to Time-Series Analysis Presented by Daniel Varadi

Introducing MUDITA: A New Approach to Time-Series Analysis PRESENTED BY DANIEL VARADI

Content overview 1. Introduction 2. Theoretical background 3. Key concepts of MUDITA 4. First experiment (descriptive learning phase) 5. Second experiment (learning and testing phases) 6. Test results 7. Conclusions / Future directions 8. References

Introduction TIME MODELLING MUDITA

Theoretical background Navie time-interpretations: Y=f(Xt) Xt=f(X1, ,Xn) Y=Xn=f(X1, ,X(n-1)) Regression models Previous own models involving time: Non-Causal Modelling Non-Causal Forecasting Yield estimation engine Xt = time!

Key concepts of MUDITA 1. Multidimensional time-layering & Similarity analyses 3. Key advantages Multidimensional Approach: Incorporates diverse time-specific attributes (e.g., years, quarters, months + arbitrary other aspects) for nuanced analysis. Similarity analyses: Utilizes advanced techniques to compare patterns across dimensions, enhancing interpretive power. 2. Flexibility & Adaptability Complex pattern capture: Captures intricate temporal patterns often overlooked by traditional methods. Improved forecasting: Enhances accuracy in forecasting by leveraging advanced analytical techniques. Custom Attributes: Allows inclusion of arbitrary time-specific attributes and custom patterns (e.g., holidays, astronomical events).

First experiment (descriptive learning phase) Raw values Ranking values for Xi Ranking values for Xi correlations types id -0.060 X1 Year 2018 2018 2018 2018 2018 2023 2023 2023 2023 2023 2023 0.438 0.354 X2 Month Day 7 7 8 8 8 6 6 6 7 7 7 0.390 0.470 1.000 X5 Quarter Week facts 3 3 3 3 3 2 2 2 3 3 3 -0.355 correlations types id 0.051 -0.437 -0.353 X1 X2 direct direct 238 110 238 110 238 238 238 1 132 1 132 1 132 1 110 1 110 1 110 -0.058 X1 inverse inverse inverse facts 1 131 1 131 1 153 1 153 1 153 233 110 233 110 233 110 233 131 233 131 233 131 0.437 0.354 12 647 0.826 correlations types id 0.051 -0.437 -0.353 0.359 X1 X2 direct direct 238 110 238 110 238 88 238 88 238 88 1 132 1 132 1 132 1 110 1 110 1 110 -0.058 X1 0.437 0.354 -0.324 (X6) 0.893 8 742 X3 X4 Y (X6) Difference(Yt-Yt-1) X3 direct X2 X3 Y Model1 Estimations ABS(Errors) 50854 58818 35875 38944 46104 58745 65321 69340 40552 43913 51512 X3 direct (X6) direct inverse inverse inverse inverse facts 93 1 131 224 1 131 7 1 153 76 1 153 112 1 153 144 233 110 89 233 110 86 233 110 21 233 131 233 233 131 83 233 131 X2 X3 Y Model2 Estimations ABS(Errors) 51803 70128 36324 36730 43335 46711 50034 61439 39400 56171 42384 1 2 3 4 5 22 29 5 12 19 11 18 25 2 9 16 29 30 31 32 33 23 24 25 26 27 28 69 83 42 35 40 66 65 62 36 53 48 0 1 2 3 4 5 74 14 219 159 99 169 109 49 245 185 125 181 241 34 94 154 86 146 206 69000 83000 42000 35000 40000 66000 65000 62000 36000 53000 48000 18146 24182 6125 3944 6104 7255 321 7340 4552 9087 3512 1 2 3 4 5 74 14 219 159 99 169 109 49 245 185 125 181 241 34 94 154 86 146 206 167 34 255 184 134 94 170 175 236 26 179 69000 83000 42000 35000 40000 66000 65000 62000 36000 53000 48000 17197 12873 5676 1730 3335 19290 14966 561 3400 3171 5616 -14 41 88 88 88 7 -5 -7 1 3 26 -17 256 257 258 259 260 261 256 257 258 259 260 261 256 257 258 259 260 261 8 8 68 128 68 128 5

Second experiment (learning and testing phases) Raw values Ranking values for Xi Input X(i) Output Regression Stair-case function Regression parameters Correlations Types Objects timeinterval1 timeinterval2 timeinterval3 timeinterval4 timeinterval5 timeinterval234 timeinterval235 timeinterval236 timeinterval237 timeinterval238 timeinterval239 ---> 0.03 -0.06 X0 Time Year Month Day Quarter Week facts Difference(Yt-Yt-1) 1 2018 7 22 3 2 2018 7 29 3 3 2018 8 5 3 4 2018 8 12 3 5 2018 8 19 3 234 2023 6 4 2 235 2023 6 11 2 236 2023 6 18 2 237 2023 6 25 2 238 2023 7 2 3 239 2023 7 16 3 Ranking values for Xi ---> ---> 0.44 X2 ---> 0.35 X3 ---> 0.39 X4 ---> 0.47 X5 ---> 1.00 Y ---> -0.38 (X6) 1 -15 0.05 X1 18 4328 24 135 0.44 X2 4453 0.35 X3 <--parameters 0.03 X0 direct direct direct direct inverse inverse inverse 1 217 102 67 2 217 102 12 3 217 81 202 4 217 81 145 5 217 81 90 234 1 122 211 235 1 122 154 236 1 122 98 237 1 122 46 238 1 102 224 239 1 102 112 -0.44 -0.35 X2 -0.05 X1 1.00 Y facst 1069000 1083000 1042000 1035000 1040000 1059000 1066000 1065000 1062000 1036000 1048000 Ranking values for Xi 0.49 16311 ABS(Errors) Estimations ABS(Errors) 25256 1067816 32401 1080302 17220 1040846 3115 1033854 3193 1038848 22436 1058077 26784 1064819 14351 1063820 4870 1065668 12080 1032355 10121 1052333 0.95 4474 X1 X3 Benchmark Regression 1043744 1050599 1024780 1031885 1043193 1036564 1039216 1050649 1057130 1023920 1037879 Model5 29 30 31 32 33 22 23 24 25 26 28 69 83 42 35 40 59 66 65 62 36 48 0 1 1 1 1 1 119 119 139 139 139 99 99 99 99 119 119 166 221 30 87 143 23 79 136 188 7 119 1184 2698 1154 1147 1152 923 1181 1180 3668 3645 4333 -14 41 7 -5 13 -7 1 3 26 -12 214 214 214 214 214 214 Input X(i) Output Regression Stair-case function Input X(i) Output Regression Stair-case function 1 -47 0.05 X1 3458 -0.44 -0.35 X2 -132 1279 0.40 (X6) -14 -0.05 X1 3564 0.44 X2 1 1195 -0.37 (X6) <--parameters 1 -112 0.05 X1 3895 -0.44 -0.35 -0.39 X2 X3 0 -904 1480 0.40 (X6) -84 -0.05 X1 4300 0.44 X2 143 0.35 X3 -1229 0.39 X4 1397 -0.37 (X6) <--parameters 0.03 X0 direct direct direct direct direct inverse inverse inverse inverse 1 217 102 67 86 2 217 102 12 201 3 217 81 202 7 4 217 81 145 69 5 217 81 90 102 234 1 122 211 62 235 1 122 154 127 236 1 122 98 82 237 1 122 46 79 238 1 102 224 23 239 1 102 112 190 0.35 X3 1.00 0.61 14465 ABS(Errors) Estimations ABS(Errors) 19526 1067962 20142 1081948 16124 1040988 91 1033995 6754 1038990 25473 1057973 42131 1064965 15907 1063967 3825 1060970 5368 1034994 11564 1046982 0.99 1729 0.03 X0 direct direct direct direct direct direct inverse inverse inverse inverse inverse 1 217 102 67 61 86 2 217 102 12 61 201 3 217 81 202 61 7 4 217 81 145 61 69 5 217 81 90 61 102 234 1 122 211 122 62 235 1 122 154 122 127 236 1 122 98 122 82 237 1 122 46 122 79 238 1 102 224 61 23 239 1 102 112 61 190 1.00 0.64 14262 ABS(Errors) Estimations ABS(Errors) 22212 1067962 22962 1081948 14273 1040988 1862 1033995 6336 1038990 19105 1057973 39221 1064965 8629 1063967 4377 1060970 10914 1034994 7025 1046982 0.99 1742 X3 Benchmark Regression 1049474 1062858 1025876 1035091 1033246 1033527 1023869 1049093 1058175 1030632 1059564 Model6 X4 Benchmark Regression 1046788 1060038 1027727 1036862 1033664 1039895 1026779 1056371 1066377 1025086 1055025 Model7 facts 1069000 1083000 1042000 1035000 1040000 1059000 1066000 1065000 1062000 1036000 1048000 facts 1069000 1083000 1042000 1035000 1040000 1059000 1066000 1065000 1062000 1036000 1048000 1 1 1 1 1 119 119 139 139 139 99 99 99 99 119 119 166 221 30 87 143 23 79 136 188 7 119 152 34 233 168 125 176 92 155 160 215 48 1038 1052 1012 1005 1010 1027 1035 1033 1030 1006 1018 1 1 1 1 1 119 119 139 139 139 99 99 99 99 119 119 166 221 30 87 143 23 79 136 188 7 119 119 119 119 119 119 60 60 60 60 119 119 152 34 233 168 125 176 92 155 160 215 48 1038 1053 1012 1005 1010 1027 1035 1033 1030 1006 1019 214 214 214 214 214 214 214 214 214 214 214 214

Second experiment (learning and testing phases) Ranking values for Xi Input X(i) Output Regression Stair-case function 1 -40 0.05 X1 -56 -0.44 -0.35 -0.39 -0.47 X2 X3 0 68 3285 1083 -17 -0.05 X1 78 0.44 X2 132 0.35 X3 -109 0.39 X4 3413 0.47 X5 1004 -0.37 (X6) <--parameters 0.03 X0 direct direct direct direct direct direct direct inverse inverse inverse inverse inverse inverse 1 217 102 67 61 107 86 2 217 102 12 61 102 201 3 217 81 202 61 97 7 4 217 81 145 61 92 69 5 217 81 90 61 87 102 234 1 122 211 122 139 62 235 1 122 154 122 134 127 236 1 122 98 122 129 82 237 1 122 46 122 124 79 238 1 102 224 61 120 23 239 1 102 112 61 112 190 0.40 (X6) 1.00 0.66 13440 ABS(Errors) Estimations ABS(Errors) 23446 1068999 23540 1082999 14604 1041998 2415 1035000 1998 1039999 25214 1058999 38159 1066000 14431 1065000 2142 1061999 9468 1035998 7445 1047999 1.00 1 X4 X5 Benchmark Regression 1045554 1059460 1027396 1037415 1038002 1033786 1027841 1050569 1059858 1026532 1055445 Model8 facts 1069000 1083000 1042000 1035000 1040000 1059000 1066000 1065000 1062000 1036000 1048000 1 1 1 1 1 119 119 139 139 139 99 99 99 99 119 119 166 221 30 87 143 23 79 136 188 7 119 119 119 119 119 119 60 60 60 60 119 119 129 134 139 144 149 97 102 107 112 117 125 152 34 233 168 125 176 92 155 160 215 48 1 1 2 0 1 1 0 0 1 2 1 214 214 214 214 214 214

Second experiment (learning and testing phases) Testdata (raw) Ranking values (Xi) Substitution values based on the learning phase Input X(i) Output Regression Stair-case function Regression parameters Correlations Types Objects timeinterval11 timeinterval42 timeinterval47 timeinterval64 timeinterval66 timeinterval71 timeinterval89 timeinterval96 timeinterval101 timeinterval103 timeinterval107 timeinterval115 timeinterval116 timeinterval120 timeinterval127 timeinterval129 timeinterval171 timeinterval179 timeinterval209 timeinterval213 timeinterval223 timeinterval238 ---> ---> ---> ---> ---> ---> ---> ---> 1 2 -15 3 X1 direct 0 201593 201593 201593 201593 372221 372221 372221 372221 372221 372221 372221 540601 540601 540601 540601 738198 738198 738198 908826 908826 908826 18 4 X2 4328 5 X3 direct 72669 38208 72669 5244 52941 11238 35211 29967 50944 31215 35710 40455 5244 15982 8990 52941 40455 52941 56437 0 38208 20727 24 6 X1 135 7 X2 inverse 0 1099 0 0 0 1099 1099 0 0 0 0 0 1099 1099 1099 1099 1099 1099 0 1099 1099 0 4453 8 X3 inverse 0 11987 0 19229 4245 19229 11987 13735 4245 13735 11987 11987 19229 16232 19229 3746 11987 4245 3746 19229 11987 13735 <--parameters 0.55 27219 ABS(Errors) Estimations 46388 36401 5204 44447 8012 10154 15655 22026 38194 45680 48115 28590 1911 39177 32611 26176 38864 37277 7279 2138 43637 20883 0.83 11488 ABS(Errors) 6585 350 11939 18591 13686 27374 19185 10574 17570 2603 2597 1756 32132 2126 5273 582 1353 5837 29446 30880 1642 10656 X0 Time Year Month Day Quarter Week Difference(Yt-Yt-1) facts 11 2018 9 30 3 42 2019 5 19 2 47 2019 6 30 2 64 2019 11 3 4 66 2019 11 24 4 71 2020 1 5 1 89 2020 5 17 2 96 2020 7 12 3 101 2020 8 23 3 103 2020 9 13 3 107 2020 10 18 4 115 2020 12 20 4 116 2021 1 3 1 120 2021 2 7 1 127 2021 4 4 2 129 2021 4 25 2 171 2022 2 20 1 179 2022 4 24 2 209 2022 11 27 4 213 2023 1 1 1 223 2023 3 19 1 238 2023 7 9 3 X1 X2 X3 X4 X5 (X6) Y X0 X1 X2 X3 X4 X5 (X6) X1 X2 X3 X4 X5 (X6) inverse 178 114 5 72 37 229 142 231 207 218 99 145 235 122 43 70 102 70 29 237 124 218 X0 Y Benchmark Regression 1090388 1072401 1085204 1070447 1091012 1051154 1072345 1072026 1085194 1077680 1085115 1091410 1051089 1057177 1058611 1079176 1071864 1080277 1099721 1052138 1074637 1073883 Modell5 direct direct direct direct direct direct direct inverse 11 217 61 6 42 169 142 90 47 169 122 6 64 169 22 218 66 169 22 53 71 124 220 202 89 124 142 105 96 124 102 145 101 124 81 59 103 124 61 138 107 124 40 98 115 124 1 84 116 76 220 218 120 76 202 186 127 76 163 211 129 76 163 46 171 27 202 84 179 27 163 53 209 27 22 29 213 1 220 234 223 1 181 90 238 1 102 170 inverse 179 98 118 218 218 20 98 138 159 179 200 239 20 38 77 77 38 77 218 20 59 138 inverse 234 150 234 22 187 38 135 95 181 102 142 156 22 54 29 194 156 187 211 6 150 70 inverse 179 118 118 239 239 59 118 179 179 179 239 239 59 59 118 118 59 118 239 59 59 179 inverse 180 91 120 203 216 4 91 128 157 171 194 235 239 23 60 73 32 73 216 239 51 124 direct 861378 725112 719619 657288 649297 609840 589946 546594 527116 500895 493902 484663 454696 432720 419235 413075 233375 210233 117936 89967 51759 999 direct 116537 58352 98058 161238 161238 0 58352 98058 110045 116537 125777 212431 0 9240 42120 42120 9240 42120 161238 0 17481 98058 inverse 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 facts 1044000 1036000 1080000 1026000 1083000 1041000 1088000 1050000 1047000 1032000 1037000 1120000 1053000 1018000 1026000 1053000 1033000 1043000 1107000 1050000 1031000 1053000 39 20 26 44 47 1 20 28 34 37 42 51 53 5 13 16 7 16 47 52 11 27 13 -7 -42 -10 -15 33 -5 36 23 26 -8 -5 47 -7 -17 -11 -8 -11 -29 49 -7 26 44 36 80 26 83 41 88 50 47 32 37 120 53 18 26 53 33 43 107 50 31 53 61 122 122 1 1 181 122 61 61 61 1 1 181 181 122 122 181 122 1 181 181 61 60 149 120 37 24 236 149 112 83 69 46 5 1 217 180 167 208 167 24 1 189 116 62 126 235 168 203 11 98 9 33 22 141 95 5 118 197 170 138 170 211 3 116 22 23 71 71 71 71 116 116 116 116 116 116 116 164 164 164 164 213 213 213 239 239 239 1050585 1036350 1091939 1044591 1069314 1013626 1068815 1060574 1064570 1034603 1039597 1121756 1020868 1015874 1031273 1053582 1034353 1048837 1077555 1019120 1029359 1042344 Substitution values based on the learning phase Substitution values based on the learning phase Input X(i) Output Regression Stair-case function Input X(i) Output Regression Stair-case function 1 2 -47 3 X1 direct 0 171353 171353 171353 171353 325521 325521 325521 325521 325521 325521 325521 435745 435745 435745 435745 578650 578650 578650 697228 697228 697228 3458 4 X2 direct 98951 47028 79837 127571 127571 0 47028 79837 96544 98951 103633 191668 0 2133 34030 34030 2133 34030 127571 0 8364 79837 -132 5 X3 direct 74086 104139 44523 74086 8891 52192 12998 142719 39516 30640 154733 49184 124225 33623 126873 44523 47481 8891 18642 11013 54089 47481 52192 56223 0 44523 26520 126873 1279 6 (X6) direct -14 7 X1 3564 8 X2 inverse 0 5915 0 0 0 5915 5915 0 0 0 0 0 5915 5915 5915 5915 5915 5915 0 5915 5915 0 1 9 1195 10 (X6) inverse 110186 143340 244945 147599 152831 61615 143340 61615 88114 88114 143340 140343 0 143340 152831 147599 143340 147599 158794 0 143340 88114 <--parameters Fehler 0.50 64406 ABS(Errors) Estimations 85720 80951 59399 90029 57988 44676 26134 55560 74264 79904 92419 11549 31664 82293 81679 74448 84248 85177 42680 34927 86724 54503 1 2 -112 3 X1 direct 0 171353 171353 171353 171353 329567 329567 329567 329567 329567 329567 329567 439791 439791 439791 439791 582696 582696 582696 701275 701275 701275 3895 4 X2 direct 98951 47028 79837 127571 127571 0 47028 79837 96544 98951 103633 191668 0 2133 34030 34030 2133 34030 127571 0 8364 79837 0 5 -904 6 X4 direct 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1480 7 (X6) direct 104139 65028 0 64042 61806 142719 65752 150687 124225 126873 65028 65752 150687 65028 61806 64042 65028 64042 53109 150687 65028 126873 -84 8 X1 4300 9 X2 inverse 0 5915 0 0 0 5915 5915 0 0 0 0 0 5915 5915 5915 5915 5915 5915 0 5915 5915 0 143 10 X3 inverse 0 16274 0 26856 14250 26856 16274 19904 14250 19585 16274 15718 26856 21891 26856 14250 15718 14250 14250 28830 16274 19904 -1229 11 X4 inverse 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1397 12 (X6) inverse 106140 139294 240899 143553 148784 61615 139294 61615 84068 84068 139294 136297 0 139294 148784 143553 139294 143553 154747 0 139294 84068 <--parameters 0.61 17882 ABS(Errors) 4880 1018 20085 8197 15962 36701 18590 11642 10610 1033 2574 2589 94295 6946 9061 2232 261 8879 37850 88721 1984 9303 0.48 22874 ABS(Errors) Estimations 36081 25684 10901 30118 303 16114 29527 8279 17570 28434 28054 40939 29562 26715 18314 12744 29387 23117 16207 26896 37968 10312 0.60 Model7 0.60 17779 ABS(Errors) 4880 1018 18911 8197 15962 32655 18590 11642 10610 714 2574 2589 94295 6946 9061 2232 261 8879 41123 88721 1984 9303 X0 X3 Y Benchmark Regression 1129720 1116951 1139399 1116029 1140988 1085676 1114134 1105560 1121264 1111904 1129419 1131549 1084664 1100293 1107679 1127448 1117248 1128177 1149680 1084927 1117724 1107503 Model6 X0 X3 Y Benchmark Estimations 1080081 1061684 1090901 1056118 1082697 1024886 1058473 1041721 1064570 1060434 1065054 1079061 1023438 1044715 1044314 1065744 1062387 1066117 1090793 1023104 1068968 1042688 direct 661518 541522 529864 487886 487036 428674 426064 389393 359771 340048 336107 336107 326566 318362 306866 295098 174475 155201 80553 74573 52312 5220 inverse 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 inverse 0 16274 0 26856 14250 26856 16274 19904 14250 19904 16274 15718 26856 21891 26856 14250 15718 14250 14250 28830 16274 19904 facts 1044000 1036000 1080000 1026000 1083000 1041000 1088000 1050000 1047000 1032000 1037000 1120000 1053000 1018000 1026000 1053000 1033000 1043000 1107000 1050000 1031000 1053000 direct 665564 545568 532736 491932 491082 428674 426064 389393 359771 340048 336107 336107 326566 318362 306866 295098 174475 155201 77281 74573 52312 5220 direct 74086 44523 74086 8891 52192 12998 39516 30640 49184 33623 44523 47481 8891 18642 11013 54089 47481 52192 56223 0 44523 26520 inverse 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 facts 1044000 1036000 1080000 1026000 1083000 1041000 1088000 1050000 1047000 1032000 1037000 1120000 1053000 1018000 1026000 1053000 1033000 1043000 1107000 1050000 1031000 1053000 1048880 1034982 1100085 1034197 1067038 1004299 1069410 1061642 1057610 1033033 1034426 1122589 958705 1011054 1035061 1050768 1032739 1051879 1069150 961279 1032984 1043697 1048880 1034982 1098911 1034197 1067038 1008345 1069410 1061642 1057610 1032714 1034426 1122589 958705 1011054 1035061 1050768 1032739 1051879 1065877 961279 1032984 1043697 65028 0 64042 61806 65752 65028 65752 154733 65028 61806 64042 65028 64042 53109 154733 65028

Second experiment (learning and testing phases) Substitution values based on the learning phase Input X(i) Output Regression Stair-case function 1 2 -40 3 X1 direct 0 181972 181972 181972 181972 317633 317633 317633 317633 317633 317633 317633 423911 423911 423911 423911 546255 546255 546255 675065 675065 675065 -56 4 X2 direct 104052 71812 71812 119745 119745 0 71812 71812 92841 104052 119745 153861 0 11211 58748 58748 11211 58748 119745 0 29377 71812 0 5 68 6 X4 3285 7 X5 direct 51247 3970 49537 81255 111146 0 3970 49537 51247 51247 51247 125195 125195 0 0 1629 0 1629 111146 125195 0 49537 1083 8 (X6) direct 50452 33716 0 33716 32118 102683 33716 104765 57798 59205 33716 33716 109716 33716 32118 33716 33716 33716 26558 123298 33716 59205 -17 9 X1 78 10 X2 132 11 X3 inverse 0 5378 0 7722 5378 7722 5378 5378 5378 5378 5378 5378 7722 5378 7722 5378 5378 5378 5378 7722 5378 5378 -109 12 X4 inverse 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3413 13 X5 inverse 4569 47853 35132 0 0 80586 47853 29498 22080 11119 0 0 0 72886 55296 55296 67611 55296 0 0 55296 33747 1004 14 (X6) inverse 131209 145552 217658 148135 156986 76464 145199 76379 122224 122224 145552 142998 0 145552 156986 148135 145552 148135 161807 0 145552 122224 <--parameters 0.59 19043 direct 30372 22035 4057 27287 3491 16794 33075 8079 14684 22835 27551 46751 294 22668 17027 11392 25033 21627 20329 2230 30387 10947 0.78 13920 Fehler 4768 2868 19617 13526 4962 29297 35039 206 15887 6021 5878 7330 66756 1893 14183 3059 27 11059 2282 48366 1023 12183 X0 X3 Y Benchmark 0 1074372 1058035 1084057 1053287 1079509 1024206 1054925 1041921 1061684 1054835 1064551 1073249 1052706 1040668 1043027 1064392 1058033 1064627 1086671 1052230 1061387 1042053 Model8 Estimations 1039232 1033132 1099617 1039526 1087962 1011703 1052961 1049794 1062887 1025979 1031122 1127330 986244 1016107 1040183 1056059 1033027 1054059 1109282 1001634 1032023 1040817 direct 642539 505616 488342 458567 435958 415618 394438 372103 353206 330743 324889 311286 311286 307591 296988 284587 186041 160243 91463 70355 50376 2644 direct 55165 37264 55165 8415 44660 10998 32963 22690 40481 24379 32963 37264 8415 15863 8415 44660 37264 44660 46931 0 37264 21206 direct 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 inverse 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 inverse 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 facts 1044000 1036000 1080000 1026000 1083000 1041000 1088000 1050000 1047000 1032000 1037000 1120000 1053000 1018000 1026000 1053000 1033000 1043000 1107000 1050000 1031000 1053000

Test results Test results 1150000 The learning accuracy generally improves with additional input attributes. Some deviations in specific cases indicate nuanced behaviour. Testing phase accuracy does not consistently mirror the learning phase trends, showcasing the challenges of independent behaviour. Including the time variable (X0) enhances the robustness of models, emphasizing its importance in improving forecasting accuracy. 1100000 1050000 1000000 950000 900000 facts Model8 Model5

Test results The accuracy of MUDITA surpasses traditional regression models, as evident in correlation values and average differences during both learning and testing phases. MUDITA outperforms regression models, demonstrating its robustness in diverse scenarios. The flexibility of MUDITA contributes to superior performance. correlations average ABS(error) Process Model learning Model1 learning Model2 learning Model3 learning Model4 learning Model5 learning Model6 learning Model7 learning Model8 testing testing testing testing regression stair-case function regression stair-case function 0.83 0.89 0.89 0.89 0.49 0.95 0.61 0.99 0.64 0.99 0.66 1.00 0.55 0.83 0.50 0.61 0.48 0.60 0.59 0.78 12647 8742 8701 9080 4474 1729 1742 16311 14465 14262 13440 27219 64406 22874 19043 1 Model5 Model6 Model7 Model8 11488 17882 17779 13920

Conculsions / Future directions Automation and online service Philosophical dilemma Ongoing research Global impact

THANK YOU! CONTACT: miau@my-x.hu Full text: https://miau.my-x.hu/miau/308/full_new_time_series_analysis_method_mudita.docx PPT: https://miau.my-x.hu/miau/308/Introducing_MUDITA.pptx

References 1. (https://miau.my-x.hu/miau/293/ncm-abstract-and-full-text-2022.pdf) (https://en.wikipedia.org/wiki/Time_series) 2. (https://miau.my-x.hu/miau/303/full_ankara_yield.pdf) 3. (https://www.google.com/search?q=dobo+andor+joker) 4. (https://miau.my-x.hu/myx-free/index.php3?x=iq) 5. (https://miau.my-x.hu/myx-free/files/iq.png) 6. (https://stellarium.org/) 7. (https://miau.my- x.hu/miau/254/coco_optimum_hatasok_std_modellekkel.xlsx)