This series of articles is devoted to the study of the construction activity of the main city of Silicon Valley — San Francisco. Charts and calculations were built with the help of Jupyter Notebook (Kaggle)

该系列文章专门研究硅谷主要城市旧金山的建筑活动。 借助Jupyter Notebook (Kaggle)构建图表和计算

Data on more than a million building permits (records in two datasets) acquired from the San Francisco Construction Department allow us to analyze not only the construction activity in the city, but also critically examine the latest trends and development history of the construction industry over the past 40 years, from 1980 to 2019 (section “Annual Construction Activity in San Francisco”).

从旧金山建筑局获得的超过一百万份建筑许可的数据(两个数据集中的记录) 使我们不仅可以分析城市的建筑活动，而且可以严谨地检查建筑业在该城市的最新趋势和发展历史。从1980年到2019年的40年(“旧金山年度建筑活动”部分)。

The movement of activity in the construction industry in San Francisco almost completely coincides with the growth schedule for gold and bitcoin (section “The future of the San Francisco construction industry, pattern prediction”)

旧金山建筑行业的活动变化几乎与黄金和比特币的增长时间表一致 (“旧金山建筑行业的未来，模式预测”部分)

Open data provides an opportunity to explore the main factors that have influenced and will have an effect on the development of the construction industry in the city, dividing them into “external” (economic booms and crises) and “internal” (the effect of holidays and seasonal-annual cycles).

开放数据为探讨影响和将影响城市建筑业发展的主要因素提供了机会，将其分为“外部”(经济繁荣和危机)和“内部”(假期的影响)和季节-年度周期)。

内容： (Content:)

1. Open data and overview of initial parameters
   打开数据并查看初始参数
2. Annual Construction Activity in San Francisco
   旧金山年度建筑活动
3. Expectation and reality in drawing up the estimated cost
   编制估计费用的期望和现实
4. Construction activity depending on the season of the year
   建筑活动取决于一年中的季节
5. Total San Francisco Real Estate Investments
   旧金山房地产投资总额
6. Areas of San Francisco that have received more investments over the past 40 years
   在过去40年中，旧金山地区获得了更多投资
7. Average estimated cost of application by city district
   按市区划分的平均估计申请费用
8. Monthly and Daily statistics on the total number of applications
   每月和每日统计的申请总数
9. The future of the San Francisco construction industry
   旧金山建筑业的未来

1.打开数据并查看初始参数。 (1. Open data and overview of initial parameters.)

San Francisco building permit data — taken from the open data portal — data.sfgov.org. The portal has several datasets on the topic of construction. Two such datasets store and update data on permits issued for the construction or repair of facilities in the city:

旧金山建筑许可数据-从开放数据门户获取-data.sfgov.org。 该门户网站具有关于建筑主题的多个数据集。 两个这样的数据集存储和更新有关为城市中的设施的建设或维修而签发的许可证的数据：

• Building permits for the period 1980–2013 (850 thousand records)

  1980-2013年的建筑许可 (85万条记录)

• Building permits for the period after 2013 (280 thousand records, data are downloaded and updated weekly)

  2013年之后的建筑许可 (28万条记录，每周下载和更新数据)

📓 These datasets contain information on the issued building permits with various characteristics of the facility for which the permit is issued. The total number of records (permits) received in the period 1980–2019 is 1,137,695 permits.

📓这些数据集包含有关已签发建筑许可证的信息，这些信息具有签发许可证的设施的各种特征。 1980-2019年期间收到的记录(许可证) 总数为1,137,695个许可证。

The main parameters from this dataset that were used for analysis:permit_creation_date - date of creation of the permit (in fact, the day from which construction work begins)desctription - description of the permit (two or three keywords describing the construction (work) object for which permission was created)estimated_cost - estimated cost of construction workrevised_cost - cost of work after revaluation, increase or decrease of the initial
volume of the applicationexisting_use - type of housing (one-, two-family house, apartments, offices, etc.)
 zipcode, location - zip code and coordinates of the object

Charts and calculations were built in the Jupyter Notebook (on the Kaggle.com platform).

图表和计算是在Jupyter Notebook (在Kaggle.com平台上)中构建的。

2.旧金山年度建筑活动 (2. Annual Construction Activity in San Francisco)

In the graph below, the data on the estimated_cost and revised_cost parameters is presented as a distribution of the total cost of work by month (in billion US dollars).

在下图中，有关估计成本和修订成本参数的数据表示为每月总工作成本的分布(十亿美元)。

data_cost_m = data_cost.groupby(pd.Grouper(freq='M')).sum()

📊 To reduce monthly “emissions”, monthly data is grouped by year. The graph of the amount of money invested over the years has received a more logical view and is amenable to analysis.

📊 为了减少每月的“排放量”，按年对每月数据进行分组 。 这些年来投资金额的图表已获得更合乎逻辑的观点，并且可以进行分析。

data_cost_y = data_cost.groupby(pd.Grouper(freq='Y')).sum()

By the annual movement of the sum of costs (all permits of the year) in urban facilities, it is seen that Economic factors from 1980 to 2019 have influenced the number and cost of construction projects or in other words, on San Francisco real estate investments.

通过城市设施成本总和的年度变动(一年中的所有许可)，可以看出1980年至2019年的经济因素已经影响了建设项目的数量和成本，换句话说，对旧金山房地产投资产生了影响。 。

The number of building permits (the number of construction works or the number of investments) over the past 40 years has been closely related to economic activity in the Silicone Valley.

在过去40年中，建筑许可的数量(建筑工程数量或投资数量)与硅谷的经济活动密切相关。

The first peak of construction activity was associated with the electronic hype of the mid-80s in the valley. The ensuing decline in electronics and banking in 1985 has led to the regional real estate market decline from which it has not yet recovered for nearly ten years.

建筑活动的第一个高峰与谷地80年代中期的电子宣传有关。 随之而来的是1985年电子和银行业务的下滑，导致该地区的房地产市场下滑，从那时起，它已经有近十年没有复苏。

🎢 Thereafter, the construction industry in San Francisco went through a parabolic growth of several thousand percent before the collapse of the Dotcom bubble and the technological boom of recent years. It happened two more times — in 1993–2000 and 2009–2016.

🎢此后，在Dotcom泡沫破灭和近年来的技术繁荣之前， 旧金山的建筑业经历了数千％的抛物线增长 。 它又发生了两次-在1993–2000年和2009–2016年。

By removing the intermediate peaks and downturns and leaving the minimum and maximum values on each economic cycle, one can see how much large market fluctuations have plagued the industry over the past 40 years.

通过消除中间的高峰和低谷，并在每个经济周期中保留最小值和最大值，您可以看到过去40年中有多少大的市场波动困扰着该行业。

The largest investment increase in the field of construction occurred during the dot-com boom, when during the period from 1993 to 2001, $ 10 billion, or about $ 1 billion a year, were invested in repairs and construction. If you count in square meters (the cost of 1 m² in 1995 is $ 3,000) — this is approximately 350,000 m2 per year for 10 years, since 1993.

建筑领域投资增长最大的时期是网络繁荣时期，当时从1993年到2001年，有100亿美元，即每年约10亿美元投资于维修和建筑。 如果以平方米计算(1995年1平方米的成本为3,000美元)-这是自1993年以来每年10年的大约350,000平方米。

The growth of annual total investments for this period amounted to 1215%.

在此期间，年度总投资增长了1215％ 。

Companies that leased construction equipment during this period were like people who sold shovels during the gold rush (in the same region in the middle of the 19th century). Only instead of shovels — in the 2000s there were already cranes and concrete pumps for the newly formed construction companies who wanted to make money on the construction boom.

在此期间租用建筑设备的公司就像在淘金热期间(在19世纪中叶的同一地区)出售铁锹的人一样。 仅用铲子代替铲子-在2000年代，已经出现了起重机和混凝土泵，供新成立的建筑公司使用，它们希望在建筑热潮中赚钱。

After each crisis that the construction industry has experienced over the years, over the next two post-crisis years, investments (the number of applications for permits) in construction fell each time by at least 50%.

在经历了建筑行业多年来的每一次危机之后，在接下来的两个危机后的几年里， 建筑业的投资(许可证的申请数量) 每次都至少下降了50％ 。

The largest crises in the construction industry in San Francisco occurred in the 90s, were with a frequency of 5 years, the industry either fell (-85% between 1983–1986), then rose again (+ 895% between 1988–1992), remaining on the same level in annual terms — 1981, 1986, 1988, 1993.

旧金山建筑业最大的危机发生在90年代，频率为5年，该行业要么下跌(1983-1986年间为-85％)，然后又上升(1988-1992年间为+ 895％)，每年保持相同水平-1981、1986、1988、1993。

🌊 After 1993, all subsequent downturns in the construction industry amounted to no more than 50%. But the approaching economic crisis (due to COVID-19) could create a record crisis in the construction industry in the period 2017–2021, the fall of which already for the period 2017–2019 amounts to more than 60%.

1993 1993年以后，建筑业随后的所有衰退不超过50％。 但是，即将到来的经济危机(由于COVID-19)可能会在2017-2021年间造成建筑业创纪录的危机，而 2017-2019年间的下降幅度已超过60％。

🏨 The population growth of San Francisco over the period 1980–1993 also showed almost exponential growth. The economic strength and innovative energy of Silicon Valley was the solid foundation upon which the hyperbole of the new economy, the American Renaissance and dotcoms was built. It was the epicenter of the new economy. But unlike the growth of real estate investments, after the peak of dotcoms, the population growth actually went to a plateau.

1980 1980年至1993年期间，旧金山的人口增长也显示了近乎指数级的增长 。 硅谷的经济实力和创新活力为新经济，美国文艺复兴时期和网络公司的夸张奠定了坚实的基础。 它是新经济的中心。 但是，与房地产投资增长不同的是，在网络泡沫破灭之后，人口增长实际上达到了一个稳定的水平。

Since the 1950s and before the peak of the dotcoms in 2001, the annual population growth has been approximately about 1% per year. Later, after a housing bubble pop included a downturn in the economy, the influx of a new population has slowed down and since 2001 it has only been 0.2 % per year.

自1950年代以来，互联网泡沫在2001年达到顶峰之前， 每年的人口增长率约为1％ 。 后来，在房地产泡沫破灭包括经济不景气之后，新人口的涌入速度有所放缓， 自2001年以来，每年的涌入率仅为0.2％ 。

In 2019 (for the first time since 1950), the growth dynamics showed an outflow of the population (-0.21% or 7000 people) from the city of San Francisco.

在2019年(自1950年以来首次)，增长动力显示旧金山市人口(-0.21％或7000人)外流。

3.编制估计费用的期望和现实 (3. Expectation and reality in drawing up the estimated cost)

In the used datasets, data on the cost of permitting a building object is divided into:

在使用的数据集中，关于建筑对象许可成本的数据分为：

• initial estimated cost (estimated_cost)

  初始估算费用( estimate_cost )

• cost of work after revaluation (revised_cost)

  重估后的工作成本( 修订成本)

🎰 During the boom, the main purpose of revaluation is to increase the initial cost, when the investor (construction customer) shows a high interest in quality and volumes after the start of construction.

the在繁荣时期，当投资者(建筑客户)在开始建造后对质量和数量表现出浓厚兴趣时，重估的主要目的是增加初始成本。

During the crisis — they tried not to exceed the estimated cost and the initial estimates , practically trying not to undergo changes (with the exception of the 1989 earthquake).

在危机期间，他们试图不超过估计的成本和最初的估计，实际上试图不进行更改(1989年地震除外)。

According to the graph of the revalued and estimated cost built on the difference (revised_cost — estimated_cost), we can observe that:

根据基于差异(revised_cost —估计成本)的重估和估计成本的图表，我们可以观察到：

The amount of cost increase during the revaluation of the volume of construction work — directly depends on the cycles of the economic boom

重估建筑工程量期间的成本增加额-直接取决于经济繁荣的周期

data_spread = data_cost.assign(spread = (data_cost.revised_cost-data_cost.estimated_cost))

During periods of rapid economic growth customers (investors) spend their money generously enough, increasing their demands after the start of work.

在经济快速增长的时期，客户(投资者)足够慷慨地花钱，开始工作后就增加了需求。

The customer (investor), feeling his financial confidence, asks the construction contractor or an architect to expand the already issued building permit. This may be a decision to increase the initial length of the pool or increase the area of the house (after the start of work and the issuance of a building permit).

客户(投资者)感到自己的财务信心，要求建筑承包商或建筑师扩大已经签发的建筑许可证。 这可能是增加游泳池的初始长度或增加房屋面积的决定(在开始工作并颁发建筑许可之后)。

At the peak of dotcoms, such “additional” expenses reached the “extra” 1 billion per year.

在互联网高峰期，这种“额外”支出每年达到“额外” 10亿美元。

If you look at this table as a percentage change, the peak increase in estimates (100% or 2 times the original estimated cost) came in the year before the earthquake in 1989 near the city. I suppose that after the earthquake (in 1989) the construction projects that were started in 1988 required more time and money to be implemented into it.

如果将此表看成是百分比变化，则估计最高峰值(100％或原始估计成本的2倍)出现在1989年地震发生前的一年。 我想在地震(1989年)之后，1988年开始的建设项目需要更多的时间和金钱来实施。

🌋 Conversely, a downward revision of the estimated cost (which happened only once during the period from 1980 to 2019) a few years before the earthquake is presumably due to the fact that some objects started in 1986–1987 were frozen or investments in these objects were cut back. According to the schedule, on average for each object begun in 1987, the estimated cost reduction was -20% of the original plan.

🌋相反，地震发生前几年的估计成本(在1980年至2019年期间仅发生过一次)的向下修订可能是由于1986-1987年开始的某些物品被冻结或对这些物品的投资被削减了。 根据时间表， 对于每个始于1987年的对象，估计的成本降低为原始计划的-20％ 。

data_spread_percent = data_cost_y.assign(spread = ((data_cost_y.revised_cost-data_cost_y.estimated_cost)/data_cost_y.estimated_cost*100))

The increase in the initial estimated cost by more than 40% indicated or possibly was the result of an approaching bubble in the financial and subsequently the construction market.

最初的估计成本增加了40％以上，这表明或可能是金融市场以及随后的建筑市场泡沫逼近的结果。

What is the reason for the decrease in the spread (difference) between the estimated and revised sum after 2007?

2007年之后的估计数和修订数之差(差异)减小的原因是什么？

Perhaps investors began to look at the numbers more carefully (the average investment over 20 years has increased from $ 100 thousand to $ 2 million dollars), or perhaps the construction department introduced new rules and restrictions to reduce possible manipulations and possible risks that arise during the crisis years in order to prevent and slow down the emerging bubbles in the real estate market.

也许投资者开始更仔细地研究数字(20年的平均投资已从10万美元增加到200万美元)，或者建筑部门出台了新的规则和限制，以减少在操作过程中可能发生的操纵和可能出现的风险危机年代是为了防止和减缓房地产市场中出现的泡沫。

4.建筑活动取决于一年中的季节 (4. Construction activity depending on the season of the year)

Having grouped the data by calendar weeks in a year (54 weeks), you can observe the construction activity of the city of San Francisco, depending on seasonality and time of year.

在一年(54周)中按日历周对数据进行分组后，您可以根据季节和一年中的时间观察旧金山市的建筑活动。

🎅 By Christmas, all construction companies are trying to manage to get permission for new “large” objects (at the same time! The number! Permits in the same months are at the same level throughout the year). Investors, planning to get their property over the next year, conclude contracts in the winter months, counting on big discounts (since summer contracts, for the most part, are coming to an end by the end of the year and construction companies are interested in receiving new applications).

Christmas 到圣诞节之前，所有建筑公司都试图设法获得新的“大型”物品的许可 (同时！数量！一年中同一月份的许可处于同一水平)。 计划在明年获得财产的投资者，在冬季月份签订了合同，依靠大幅度的折扣(因为夏季合同大部分将在年底到期，而建筑公司对接收新的申请)。

Before Christmas, the largest amounts are submitted in applications (an increase from an average of 1–1.5 billion per month. Up to 5 billion in December alone). At the same time, the total number of applications by month remains at the same level (see the section below: Statistics on the total number of applications by month and days)

圣诞节之前，提交的申请数量最多 (从平均每月1–15亿增加到12月份的50亿)。 同时，每月申请总数保持在相同水平 (请参阅以下部分：按月份和天数统计的申请总数)

After the winter holidays, the construction industry is actively (almost without an increase in the number of permits) planning and implementing “Christmas” orders, so that by the middle of the year (before the Independence Day) have time to free up resources before the beginning of immediately after the June holidays — a new wave of summer agreements.

寒假过后，建筑行业正在积极(几乎没有增加许可证的数量)计划和实施“圣诞节”命令，以便在年中(独立日之前)有时间释放资源六月假期后立即开始-夏季协议的新潮。

data_month_year = data_month_year.assign(week_year = data_month_year.permit_creation_date.dt.week)data_month_year = data_month_year.groupby(['week_year'])['estimated_cost'].sum()

The same percentage data (orange line) also shows that the industry works “quietly” for a year, but before and after the holidays, permit activity increases to 150% between week 20–24 (before Independence Day), and decreases immediately after the holiday to -70%.

相同的百分比数据(橙色线)还表明，该行业“安静”地工作了一年，但是在假期之前和之后，许可证活动在20-24周(独立日之前)增加到150％，而在假期之后立即减少。假期到-70％。

Before Halloween and Christmas, activity in the construction industry in San Francisco week 43–44 increases by 150% (from bottom to peak) and then decreases to zero during the holidays.

在万圣节和圣诞节之前，旧金山第43-44周的建筑业活动增加了150％(从底部到高峰)，然后在假日期间减少到零。

Therefore, the construction industry is in a six-month cycle, which is divided by the holidays “Independence Day of the USA” (week 20) and “Christmas” (week 52).

因此， 建筑行业以六个月为周期 ，除以假期“美国独立日”(第20周)和“圣诞节”(第52周)。

5.旧金山房地产投资总额 (5. Total San Francisco Real Estate Investments)

Based on the data on building permits in the city:

根据城市建筑许可的数据：

The total investment in construction projects in San Francisco from 1980 to 2019 is $ 91.5 billion.

1980年至2019年，旧金山建设项目的总投资 为915亿美元 。

sf_worth = data_location_lang_long.cost.sum()

The total market value of all residential real estate in San Francisco, estimated by property tax (is the estimated value of all real estate and all personal property owned by San Francisco) has reached $ 208 billion in 2016.

通过财产税估算的旧金山所有住宅房地产的总市值 (即旧金山拥有的所有房地产和所有个人财产的估算价值) 在2016年 已 达到2080亿美元 。

6.在过去的40年中，旧金山在哪些地区进行了更多投资 (6. In which areas of San Francisco have invested more over the past 40 years)

With the help of the Folium library, let’s see where these $ 91.5 billion by regions were invested. To do this, grouping the data by zip code (zipcode), imagine the value obtained using circles (Circle function from the Folium library).

在Folium库的帮助下，让我们了解一下这915亿美元的地区投资额。 为此，按邮政编码(zipcode)对数据进行分组，想象一下使用圆获得的值(来自Folium库的Circle函数)。

import folium
from folium import Circle
from folium import Marker
from folium.features import DivIcon# map folium display
lat = data_location_lang_long.lat.mean()
long = data_location_lang_long.long.mean()
map1 = folium.Map(location = [lat, long], zoom_start = 12)for i in range(0,len(data_location_lang_long)):
    Circle(
        location = [data_location_lang_long.iloc[i]['lat'], data_location_lang_long.iloc[i]['long']],
        radius= [data_location_lang_long.iloc[i]['cost']/20000000],
        fill = True, fill_color='#cc0000',color='#cc0000').add_to(map1)
    Marker(
    [data_location_mean.iloc[i]['lat'], data_location_mean.iloc[i]['long']],
    icon=DivIcon(
        icon_size=(6000,3336),
        icon_anchor=(0,0),
        html='<div style="font-size: 14pt; text-shadow: 0 0 10px #fff, 0 0 10px #fff;; color: #000";"">%s</div>'
        %("$ "+ str((data_location_lang_long.iloc[i]['cost']/1000000000).round()) + ' mlrd.'))).add_to(map1)
map1

By looking at districts, it becomes clear that the majority of investments went to DownTown. Having simplified the grouping of all objects according to the distance to the city center and the time needed to get to the city center (of course, expensive houses are also being built on the coast), all permissions were divided into 4 groups: ‘Downtown’, ‘<0.5H Downtown’, ‘< 1H Downtown ‘,’ Outside SF ‘.

通过查看地区，很明显，大部分投资都投向了DownTown 。 根据到市中心的距离和到达市中心所需的时间简化了所有对象的分组(当然，沿海地区也正在建造昂贵的房屋)，所有权限分为4组： ”，“ <0.5H市区”，“ <1H市区”，“ SF外”。

from geopy.distance import vincenty
def distance_calc (row):
    start = (row['lat'], row['long'])
    stop = (37.7945742, -122.3999445)    return vincenty(start, stop).meters/1000df_pr['distance'] = df_pr.apply (lambda row: distance_calc (row),axis=1)def downtown_proximity(dist):
    '''
    < 2 -> Near Downtown,  >= 2, <4 -> <0.5H Downtown
    >= 4, <6 -> <1H Downtown, >= 8 -> Outside SF
    '''
    if dist < 2:
        return 'Downtown'
    elif dist < 4:
        return  '<0.5H Downtown'
    elif dist < 6:
        return '<1H Downtown'
    elif dist >= 6:
        return 'Outside SF'
df_pr['downtown_proximity'] = df_pr.distance.apply(downtown_proximity)

91.5 billion that were invested in the city, almost 70 billion (75% of all investments) are invested in repairs and construction in the city center (green zone) and in the city area within a 2 km radius from the center (blue zone).

在该市投资了915亿美元，其中将近700亿(占投资总额的75％)在市中心 (绿色区域)和距市中心2公里半径内的城市区域(蓝色区域) 进行了维修和建设。 。

7.按市区划分的建筑申请平均估计费用 (7. Average estimated cost of an application for construction by city district)

All data, as in the case of the total amount of investments, was grouped by zip code. Only in this case with the average (.mean ()) estimated cost of the application by zip code.

与总投资额一样，所有数据均按邮政编码分组。 仅在这种情况下，使用邮政编码估算应用程序的平均成本(.mean())。

data_location_mean = data_location.groupby(['zipcode'])['lat','long','estimated_cost'].mean()

In ordinary areas of the city (more than 2 km. From the city center) — the average estimated cost of an application for construction is $ 50 thousand.

在城市普通地区(距市中心2公里以上)，平均每份施工申请的估计费用为5万美元。

The average estimated cost in the area of the city center is about three times higher ($ 150 thousand to $ 400 thousand) than in other areas ($ 30–50 thousand).

市中心区域 (15万至40万美元)的平均估计成本约为其他地区(3万至5万美元)的三倍。

In addition to the cost of land, three factors determine the total cost of housing construction: labor, materials, and government fees. These three components are higher in California than in the rest of the country. California building codes are considered among the most comprehensive and stringent in the country (due to earthquakes and environmental regulations), often requiring more expensive materials and labor.

除了土地成本外，三个因素还决定了房屋建设的总成本：人工，材料和政府费用。 这三个组成部分在加利福尼亚州高于全国其他地区。 由于地震和环境法规的原因， 加利福尼亚州的建筑法规被认为是美国最全面，最严格的法规，通常需要更昂贵的材料和劳动力。

For example, the State requires builders to use higher quality building materials (windows, insulation, heating and cooling systems) to achieve high standards in energy efficiency.

例如，纽约州要求建筑商使用更高质量的建筑材料(窗户，隔热，加热和冷却系统)以达到高能效标准。

From the general statistics on the average cost of an application for permission, two locations stand out favorably:

从有关许可平均费用的一般统计数据来看，有两个地方比较突出：

• Treasure Island — is an artificial island in the San Francisco Bay. The average estimated cost of a building permit is $ 6.5 million.

  金银岛 -是旧金山湾的人工岛。 建筑许可的平均估计费用为650万美元。

• Mission Bay — (lives 2926 people) The average estimated cost of a building permit is $ 1.5 million.

  米森湾 ( Mission Bay) -(可容纳2926人)建筑许可证的平均估计成本为150万美元。

In fact, the highest average claim in these two areas is associated with the lowest number of applications for this zip code (145 and 3064 respectively, construction on the island is very limited), while for the rest of the postal codes for the period 1980–2019, approximately 1300 applications were received per year (total average of 30–50 thousand applications for the entire period).

实际上， 这两个地区的平均索赔额最高，与此邮政编码的申请数量最少 (分别为145和3064，岛上的建筑非常有限)，而1980年其余的邮政编码–2019年，每年大约收到1300份申请(整个期间平均平均35,000份申请)。

By the parameter “number of permits” is noticeable a perfectly even distribution of the number of applications per zip code throughout the city.

通过参数“许可证​​数量”可以明显看出，整个城市每个邮政编码的申请数量分布非常均匀 。

8.按月和日统计的申请总数 (8. Statistics on the total number of applications by month and day)

General statistics on the number of applications by month and day from 1980 to 2019 shows that the “quietest” months for construction departments — are spring and winter months. At the same time, the amount of investments offered in the applications varies greatly, and it differs from month to month (see “Construction activity depending on the season of the year”). Among the days of the week on Monday, the department’s workload is approximately 20% less than the rest of the week.

根据1980年至2019年每月和每天的申请数量的一般统计数据显示， 建筑部门“最安静”的月份是Spring和冬季 。 同时， 应用程序中提供的投资金额差异很大，并且每个月都不相同 (请参阅“建筑活动取决于一年中的季节”)。 在星期一的一周中的几天中，部门的工作量比一周中的其余时间减少大约20％。

data_month_count  = data_month.groupby(['permit_creation_date']).count()

While June and July practically do not differ in the number of applications, the difference in total estimated cost reaches 100% (4.3 billion in May and July and 8.2 billion in June).

虽然6月和7月的申请数量实际上没有差异，但总估算成本的差异达到了100％(5月和7月为43亿，6月为82亿)。

data_month_sum  = data_month.groupby(['permit_creation_date']).sum()

9.旧金山建筑业的未来，模式预测。 (9. The future of the San Francisco construction industry, pattern prediction.)

In conclusion, we compare the graph of construction activity in San Francisco with the graph of Bitcoin prices (2015–2018) and the graph of gold prices (1940–1980)

总之，我们将旧金山的建筑活动图与比特币价格图(2015–2018)和黄金价格图(1940–1980)进行了比较。

Pattern — in technical analysis is a stable repeated combinations of price, volume or indicator data. Pattern analysis is based on one of the axioms of technical analysis: “history repeats itself” — it is believed that repeated combinations of data lead to a similar result. Technical analysts have long used price patterns to examine current movements and forecast future market movements.

模式 -技术分析中是价格，数量或指标数据的稳定重复组合。 模式分析基于技术分析的公理之一：“历史重复本身” —相信重复的数据组合会产生相似的结果。 技术分析师长期以来一直使用价格模式来检查当前走势并预测未来市场走势。

📈📉 Economic patterns have changed little from the ancient past to recent times. The main pattern that can be guessed on the annual activity chart is “Head and shoulders” — a trend reversal pattern. It is named because the graph looks like a human head (peak) and shoulders on the sides (smaller peaks). When the price breaks the line connecting the troughs, the pattern is considered complete, and the movement is likely to occur down.

📈📉 从古代到最近，经济格局几乎没有变化。 年度活动图表上可以猜到的主要模式是“头和肩膀” ， 即趋势反转模式 。 之所以命名，是因为该图看起来像人的头(峰值)，而肩膀在侧面(较小的峰)。 当价格跌破连接谷底的线时，该形态被认为是完整的，并且移动很可能发生。

The movement of activity in the construction industry in San Francisco almost completely coincides with the growth schedule for gold and bitcoin. The historical indicators of these three graphs of price and activity movement show significant similarities.

旧金山建筑业的活动变动与黄金和比特币的增长时间表几乎完全吻合 。 这三个价格和活动变动图的历史指标显示出明显的相似性。

In the future, it is necessary to calculate the correlation coefficient with each of these two trends. Two random variables are called correlated if their correlation moment (or correlation coefficient) is nonzero, and are called uncorrelated quantities if their correlation moment is zero. If the obtained value is closer to 0 than to 1, then talking about a clear pattern does not make sense. This is a difficult mathematical problem, which senior comrades may possibly take on, who may be interested in this topic.

将来，有必要计算这两个趋势的相关系数 。 如果两个随机变量的相关矩(或相关系数)不为零，则称为相关变量；如果两个相关变量的相关矩为零，则称为不相关量。 如果获得的值更接近于0而不是接近于1，则谈论清晰的模式是没有意义的。 这是一个很难解决的数学问题，对此同志可能会感兴趣的高级同志可能会遇到。

🔮 !Unscientific! we can look at the topic of further development of the San Francisco construction industry through the similarity of patterns. If the pattern matches further with the price of bitcoin, then according to this pessimistic option — coming out of the crisis in the construction industry in San Francisco will not be easy for the near post-crisis time.

Un ！不科学！ 我们可以通过模式的相似性来探讨旧金山建筑业的进一步发展主题。 如果这种模式与比特币的价格进一步匹配，那么根据这种悲观的选择 ，在危机后的近段时间内摆脱旧金山建筑行业的危机将并不容易。

With a more “optimistic” development option, a repeated exponential growth of the construction industry is possible if activity here goes according to the “gold price” scenario. In this option, in 20–30 years (maybe in 10), the construction sector expects a new surge in employment and development.

如果采用“更乐观”的发展选择 ，那么根据“黄金价格”情景进行的活动可以使建筑业实现指数级增长。 在这种选择下，建筑业预计在20至30年内(可能在10年内)将出现就业和发展的新趋势。

In the next part, I will take a closer look at individual sectors of construction (repair of roofs, kitchens, construction of stairs, bathrooms, and if you wish — for industries or other data; please leave me a comment) and compare inflation for individual types of work with Fixed Mortgage Rates & US Treasury Yield.

在下一部分中，我将仔细研究各个建筑部门(修理屋顶，厨房，楼梯，浴室的结构，如果需要，请提供行业或其他数据；请给我评论)，并比较具有固定抵押贷款利率和美国国债收益率的各种类型的工作。

Link to Jupyter Notebook: San Francisco. Building sector 1980–2019.

链接到Jupyter Notebook： 旧金山。 建筑业1980–2019年 。

Please, those who are registered on Kaggle — put a plus to this Notebook (Thank you!)(Notebooks will later add code comments and explanations)

请在Kaggle上注册的用户在此笔记本上加一个号(谢谢！)(笔记本将在以后添加代码注释和解释)

☕️ If you like my content, please consider buying me a coffee. Thank you for your support, I really appreciate it! buymeacoffee.com/boikoartem

☕️如果您喜欢我的内容，请考虑给我买一杯咖啡。 感谢您的支持，我真的很感激！ buymeacoffee.com/boikoartem

📈 More about various tools for working with big data visualization here:

here此处提供有关使用大数据可视化的各种工具的更多信息：

可视化。 大数据可视化工具 (Visualization. Big Data Visualization Tools)

You can learn more about working with Jupyter Notebook and about applying machine learning in construction:

您可以了解有关使用Jupyter Notebook以及在建筑中应用机器学习的更多信息：

价格和时间预测。 机器学习。 (Price and Time Prediction. Machine Learning.)