Untitled Python Project

import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns

df = pd.read_csv('/work/tord_v3_edited.csv') df.info()

df.head(20)

df.describe()

# sanitize "raised_usd" df['raised_usd'] = df['raised_usd'].fillna(0) # create outcome variable success df['success'] = np.where(df['raised_usd'] >= 500000, 1, 0)

df['success'].describe()

sns.pairplot(df, vars = ['raised_usd','teamsize','success'])

# Potential multicollinearity issue corr = df.corr() sns.heatmap(corr) plt.show()

!pip install statsmodels

import statsmodels.api as sm from statsmodels.formula.api import ols, logit # dependent/target/outcome variable y = df['success'] # independent/predictor/explanatory variable X = df[['teamsize', 'rating']] # A. Logit regression # turn independent variables into floating type (best practice) # 'missing='drop'' drops rows with missing values from the regression logit_model=sm.Logit(y,X.astype(float), missing='drop' ) # fit logit model into the data result=logit_model.fit() # summarize the logit model print(result.summary2())

sns.regplot(x = "rating", y = "success", data = df, logistic = True, y_jitter = .05) plt.ylabel("success probability")

''' options for "at" 1. 'overall' The average of the marginal effects at each observation 2. 'mean' The marginal effects at the mean of each regressor 3. 'median' The marginal effects at the median of each regressor 4. 'zero' The marginal effects at zero for each regressor 5. 'all' The marginal effects at each observation. options for "method" 1. 'dydx' No transformation is made and amrginal effects are returned 2. 'eyex' estimate elasticities of variables in exog 3. 'dyex' estimate semi-elasticity 4. 'eydx' estimate semi-elasticity ''' average_marginal_effect = result.get_margeff(at = "mean", method = "dydx") print(average_marginal_effect.summary())

# B. Linear Probability Model # logit regression X = sm.add_constant(X) ols_model=sm.OLS(y,X.astype(float), missing='drop') result=ols_model.fit() print(result.summary2())

# Use wrapper lazypredict !pip install lazypredict

from lazypredict.Supervised import LazyClassifier, LazyRegressor from sklearn.model_selection import train_test_split

# load data X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.2, random_state=42) # fit all models clf = LazyClassifier(predictions=True) models, predictions = clf.fit(X_train, X_test, y_train, y_test)

models

predictions

success = np.where(df['raised_usd'] > 500000, 1, 0) from matplotlib import pyplot

amount = df['token_for_sale'].fillna(0) sns.regplot(x = amount, y = success, data = df, logistic = True, y_jitter = .05) plt.ylabel("success probability") pyplot.xscale('log')

accepting = df['accepting'].str.count(",") + 1 sns.regplot(x = accepting, y = success, data = df, logistic = True, y_jitter = .05) plt.ylabel("success probability")

areas = df['restricted_areas'].str.count(",") + 1 sns.regplot(x = areas, y = success, data = df, logistic = True, y_jitter = .05) plt.ylabel("success probability")

sns.regplot(x = "teamsize", y = success, data = df, logistic = True, y_jitter = .05) plt.ylabel("success probability")

Q2: While keeping the Y-axis constant (the success probability), we can take a look at different X's that might suggest whether or not a coin will be successful. From the graphs above, we can see that as the rating goes higher, the success probability goes higher. Similarly, as the token for sale, accepting currencies, and team size gets larger, the success probability of a coin goes up. Conversely, as the number of restricted areas is higher, the success probability drops because less people can adopt the coin.

Q3: Variables that may be used from the data set to find meaningful X could be the country, is_ico and raised_usd, with success in Y, since such factors may help identify the performances of previous ICOs according to success and provide a guidance of successful and unsuccessful upcoming ICOs. Additionally, the heatmap helps visualize the different correlations between variables to see whether the data is meaningful and which combinations show a greater importance, hence providing a tool to make decisions on which variables should stand in X to be meaningful.

Q4: Firstly, we can use ‘is_ico’ to determine whether the data fits the scope, whether the offer is a subject for our study field. Secondly, we can compare the 'raised_usd' set with 500,000 to determine the success of the ICO. For a successful ICO, it should raise more than this amount. For countries, we can determine the success of different ICO in different countries based on the combination of ‘raised_usd’ and ‘country’ -the success of different ICO in different countries will be different due to different national policies, promotion degree and public opinion. So we might be able to find some possible connections by using the combination of ‘raised_usd’ and ‘country’.