Hanyar Mafi Sauƙi: Gabatarwa da Amfani a Nazarin Bayanai
Pendahuluan
Hanyar Mafi Sauƙi (Least Squares) tana ɗaya daga cikin dabarun da aka fi amfani da su a fannin nazarin bayanai, musamman a cikin kididdiga da lissafi da aka yi amfani da su. Wannan hanyar tana da nufin ƙiyasta sigogi waɗanda ke rage jimlar murabba'in karkacewar da aka lura daga samfurin da aka gabatar. A cikin wannan labarin, za mu bincika mahimman ra'ayoyin hanyar mafi ƙarancin murabba'i, aikace-aikacenta a fannoni daban-daban, da kuma matakai masu amfani don aiwatar da ita.
Ka'idoji na Asali na Hanyar Mafi Karancin Murabba'i
Za a iya bayyana hanyar Mafi Girman Murabba'i ta hanyar komawa kan layi. A ce muna da bayanai a cikin nau'i-nau'i \((x_i, y_i)\) inda \(i = 1, 2, …, n \). Za a iya bayyana samfurin layi da muke son ginawa kamar haka:
\[ y = \beta_0 + \beta_1 x + \epsilon \]
inda \( \beta_0 \) da \( \beta_1 \) su ne sigogin da muke son kimantawa, yayin da \( \epsilon \) shine kuskuren ko ragowar da ake tsammanin yana da matsakaicin sifili.
Manufar hanyar mafi ƙarancin murabba'i ita ce rage aikin da ke tafe:
\[ S(\beta_0, \beta_1) = \sum_{i=1}^{n} (y_i – \beta_0 – \beta_1 x_i)^2 \]
Nemo Ma'auni Mafi Kyau: Hanyar Lissafi
Domin nemo ƙimar sigogi waɗanda ke rage aikin manufa \(S \), muna buƙatar ƙididdige ɓangarorin da suka samo asali na \(S \) dangane da \( \beta_0 \) da \( \beta_1 \), sannan mu warware lissafin mai zuwa:
\[ \frac{\partial S}{\partial \beta_0} = -2 \sum_{i=1}^n (y_i – \beta_0 – \beta_1 x_i) = 0 \]
\[ \frac{\partial S}{\partial \beta_1} = -2 \sum_{i=1}^n x_i (y_i – \beta_0 – \beta_1 x_i) = 0 \]
Ta hanyar warware wannan tsarin lissafin layi, za mu iya samun masu kimantawa \(\hat{\beta_0}\) da \(\hat{\beta_1}\):
\[ \hat{\beta_1} = \frac{n \sum_{i=1}^n x_i y_i – \sum_{i=1}^n x_i \sum_{i=1}^n y_i}{n \sum_{i=1}^n x_i^2 – (\sum_{i=1}^n x_i)^2} \]
\[ \hat{\beta_0} = \bar{y} – \hat{\beta_1} \bar{x} \]
inda \(\bar{y}\) da \(\bar{x}\) sune matsakaicin \(y\) da \(x\) bi da bi.
Amfani da Hanyar Mafi Sauƙi
1. Tattalin Arziki da Kuɗi
Ana amfani da hanyar mafi ƙarancin murabba'i sosai a fannin tattalin arziki don yin kwaikwayon alaƙa tsakanin masu canjin tattalin arziki. Misali, mai sharhi kan tattalin arziki zai iya son yin kwaikwayon tasirin rashin aikin yi akan hauhawar farashi. Ta amfani da hanyar mafi ƙarancin murabba'i, mai sharhi zai iya ƙirƙirar samfurin koma-baya wanda ya danganta masu canjin biyu kuma ya yi hasashen ƙididdiga game da ƙarfi da yanayin dangantakar.
2. Kimiyyar Zamantakewa
A cikin ilimin zamantakewa, ana amfani da hanyar mafi ƙarancin murabba'i a cikin bincike da binciken tunani don nazarin alaƙar da ke tsakanin halayen ɗan adam da sauran masu canji. Misali na yau da kullun shine koma-baya mai sauƙi na layi wanda ya danganta matakin farin cikin mutum da kuɗin shigarsa na shekara-shekara.
3. Injiniyanci
A fannin injiniyanci, ana iya amfani da hanyar mafi ƙarancin murabba'i don daidaita kayan aiki da sarrafa sigina. Misali, a fannin sarrafa hotuna na dijital, ana amfani da wannan hanyar don rage hayaniya a cikin hotuna ta hanyar sanya samfurin bisa ga bayanan da aka lura.
4. Ilimin Yanayi da Yanayi
Masana yanayi suna amfani da wannan hanyar don yin nazarin bayanai da suka shafi yanayin zafi, ruwan sama, ko wasu masu canjin yanayi. Tare da samfuran komawa baya, suna iya hasashen yanayin yanayi bisa ga bayanan tarihi, suna taimakawa wajen ƙirƙirar hasashen da ya fi daidai.
Aiwatarwa Mai Amfani da Python
Domin aiwatar da hanyar da ba ta kai girman murabba'i ba a aikace, musamman ma mai sauƙin komawa ga layi, za mu iya amfani da harshen shirye-shiryen Python tare da taimakon ɗakunan karatu na 'numpy' da 'matplotlib'. Ga misalin lambar da ke nuna wannan tsari:
"' Python
shigo da numpy as np
shigo da matplotlib.pyplot azaman plt
Samfurin bayanai
x = np.array([1, 2, 3, 4, 5])
y = np.array ([2, 3, 5, 7, 11])
Matsakaicin x da y
ma'ana_x = np.ma'ana(x)
mean_y = np.matsakaicin(y)
Lissafi sigogi
Mai ƙidaya = np.sum((x – mean_x) (y – mean_y))
ma'aunin = np.sum((x – ma'ana_x) 2)
b1 = mai ƙidaya / mai ƙidaya
b0 = ma'ana_y – b1 ma'ana_x
Hasashen y
y_pred = b0 + b1 x
Sakamakon labarin
plt.scatter(x, y, launi='shuɗi', lakabi='Bayanan Lura')
plt.plot(x, y_pred, launi='ja', label='Layin Komawa')
plt.xlabel('x')
alamar plt.ylabel('y')
labari ()
plt.ya nuna ()
buga(f”Ma'aunin komawa baya: b0 = {b0}, b1 = {b1}”)
““
Kammalawa
Hanyar ƙananan murabba'ai tushe ne mai ƙarfi kuma mai mahimmanci a cikin kididdiga da nazarin bayanai. Ikonsa na rage kurakurai da haɓaka dacewa da samfura ya sa ya zama mai matuƙar amfani a fannoni daban-daban, tun daga tattalin arziki zuwa injiniyanci da kimiyyar zamantakewa. Duk da cewa manufar asali mai sauƙi ce, ana iya faɗaɗa hanyar zuwa samfuran da suka fi rikitarwa kamar koma-baya mara layi, samfuran tasirin gauraye, da koyon injina. Tare da fahimtar hanyar ƙananan murabba'ai da isasshen aiki, za mu iya inganta daidaiton nazarin bayananmu da kuma yanke shawara mai zurfi.
Da fatan wannan labarin zai samar da cikakken bayani game da hanyar da ba ta da ma'auni da kuma aikace-aikacenta.
